WhitzardNUWA Lab ↗
NUWANUWA INDEX

Search NUWA Index

Research · Benchmarks · Findings · Perspectives

中文

RESEARCH RECORD

SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents

Sheng Yin · Xianghe Pang · Yuanzhuo Ding · Menglan Chen · Yutong Bi · Yichen Xiong · Wenhao Huang · Zhen Xiang · Jing Shao · Siheng Chen · Shanghai Jiao Tong University; University of Georgia; Shanghai AI Laboratory

Original publication ↗ · BibTeX · RIS

The revised paper reports weak hazard awareness across planning approaches, despite differences in task completion. Changing the underlying LLM alone does not reliably solve the problem.

Research question

Can embodied planners recognize hazardous instructions while completing safe tasks?

Design & assumptions

Executable household simulations distinguish detailed, abstract, and long-horizon tasks; execution and semantic judgments provide complementary measurements.

Findings

The revised paper reports weak hazard awareness across planning approaches, despite differences in task completion. Changing the underlying LLM alone does not reliably solve the problem.

Limits & counterevidence

The low-level controller maps high-level actions to simulator APIs; this is not validation of a physical robot or learned motor controller. Hazardous examples are not reproduced here.

Source evidence

SystemMeasureReported valueConditions & location
GPT-4 / Lota-BenchRejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / Lota-BenchRisk ratio (goal evaluator)0.60ratioDetailed hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / Lota-BenchRisk ratio (LLM evaluator)0.38ratioDetailed hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / Lota-BenchExecution ratio0.89ratioDetailed hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / Lota-BenchExecution time20.78sDetailed hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Detailed hazardous tasks; column Execution time
GPT-4 / Lota-BenchRejection ratio0.00ratioAbstract hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / Lota-BenchRisk ratio0.59ratioAbstract hazardous tasks#S4.T2; row GPT-4 / Lota-Bench; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / Lota-BenchCompleted safe ratio0.49ratioLong-horizon tasks#S4.T2; row GPT-4 / Lota-Bench; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / Lota-BenchCompleted unsafe ratio0.28ratioLong-horizon tasks#S4.T2; row GPT-4 / Lota-Bench; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / Lota-BenchIncomplete ratio0.23ratioLong-horizon tasks#S4.T2; row GPT-4 / Lota-Bench; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / LLM-PlannerRejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / LLM-PlannerRisk ratio (goal evaluator)0.40ratioDetailed hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / LLM-PlannerRisk ratio (LLM evaluator)0.46ratioDetailed hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / LLM-PlannerExecution ratio0.75ratioDetailed hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / LLM-PlannerExecution time58.75sDetailed hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Detailed hazardous tasks; column Execution time
GPT-4 / LLM-PlannerRejection ratio0.31ratioAbstract hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / LLM-PlannerRisk ratio0.32ratioAbstract hazardous tasks#S4.T2; row GPT-4 / LLM-Planner; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / LLM-PlannerCompleted safe ratio0.14ratioLong-horizon tasks#S4.T2; row GPT-4 / LLM-Planner; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / LLM-PlannerCompleted unsafe ratio0.14ratioLong-horizon tasks#S4.T2; row GPT-4 / LLM-Planner; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / LLM-PlannerIncomplete ratio0.72ratioLong-horizon tasks#S4.T2; row GPT-4 / LLM-Planner; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / CoELARejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / CoELARisk ratio (goal evaluator)0.16ratioDetailed hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / CoELARisk ratio (LLM evaluator)0.09ratioDetailed hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / CoELAExecution ratio0.33ratioDetailed hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / CoELAExecution time74.12sDetailed hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Detailed hazardous tasks; column Execution time
GPT-4 / CoELARejection ratio0.00ratioAbstract hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / CoELARisk ratio0.21ratioAbstract hazardous tasks#S4.T2; row GPT-4 / CoELA; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / CoELACompleted safe ratio0.02ratioLong-horizon tasks#S4.T2; row GPT-4 / CoELA; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / CoELACompleted unsafe ratio0.02ratioLong-horizon tasks#S4.T2; row GPT-4 / CoELA; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / CoELAIncomplete ratio0.96ratioLong-horizon tasks#S4.T2; row GPT-4 / CoELA; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / MLDTRejection ratio0.05ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / MLDTRisk ratio (goal evaluator)0.54ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / MLDTRisk ratio (LLM evaluator)0.69ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / MLDTExecution ratio0.73ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / MLDTExecution time31.92sDetailed hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Detailed hazardous tasks; column Execution time
GPT-4 / MLDTRejection ratio0.15ratioAbstract hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / MLDTRisk ratio0.40ratioAbstract hazardous tasks#S4.T2; row GPT-4 / MLDT; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / MLDTCompleted safe ratio0.49ratioLong-horizon tasks#S4.T2; row GPT-4 / MLDT; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / MLDTCompleted unsafe ratio0.28ratioLong-horizon tasks#S4.T2; row GPT-4 / MLDT; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / MLDTIncomplete ratio0.23ratioLong-horizon tasks#S4.T2; row GPT-4 / MLDT; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / ProgPromptRejection ratio0.07ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / ProgPromptRisk ratio (goal evaluator)0.51ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / ProgPromptRisk ratio (LLM evaluator)0.68ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / ProgPromptExecution ratio0.30ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / ProgPromptExecution time22.98sDetailed hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Detailed hazardous tasks; column Execution time
GPT-4 / ProgPromptRejection ratio0.20ratioAbstract hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / ProgPromptRisk ratio0.40ratioAbstract hazardous tasks#S4.T2; row GPT-4 / ProgPrompt; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / ProgPromptCompleted safe ratio0.50ratioLong-horizon tasks#S4.T2; row GPT-4 / ProgPrompt; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / ProgPromptCompleted unsafe ratio0.23ratioLong-horizon tasks#S4.T2; row GPT-4 / ProgPrompt; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / ProgPromptIncomplete ratio0.27ratioLong-horizon tasks#S4.T2; row GPT-4 / ProgPrompt; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / MAPRejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MAP; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / MAPRisk ratio (goal evaluator)0.27ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MAP; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / MAPRisk ratio (LLM evaluator)0.31ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MAP; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / MAPExecution ratio0.64ratioDetailed hazardous tasks#S4.T2; row GPT-4 / MAP; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / MAPExecution time23.56sDetailed hazardous tasks#S4.T2; row GPT-4 / MAP; condition Detailed hazardous tasks; column Execution time
GPT-4 / MAPRejection ratio0.00ratioAbstract hazardous tasks#S4.T2; row GPT-4 / MAP; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / MAPRisk ratio0.29ratioAbstract hazardous tasks#S4.T2; row GPT-4 / MAP; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / MAPCompleted safe ratio0.23ratioLong-horizon tasks#S4.T2; row GPT-4 / MAP; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / MAPCompleted unsafe ratio0.25ratioLong-horizon tasks#S4.T2; row GPT-4 / MAP; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / MAPIncomplete ratio0.52ratioLong-horizon tasks#S4.T2; row GPT-4 / MAP; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / ReActRejection ratio0.10ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / ReActRisk ratio (goal evaluator)0.42ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / ReActRisk ratio (LLM evaluator)0.48ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / ReActExecution ratio0.74ratioDetailed hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / ReActExecution time26.95sDetailed hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Detailed hazardous tasks; column Execution time
GPT-4 / ReActRejection ratio0.32ratioAbstract hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / ReActRisk ratio0.55ratioAbstract hazardous tasks#S4.T2; row GPT-4 / ReAct; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / ReActCompleted safe ratio0.04ratioLong-horizon tasks#S4.T2; row GPT-4 / ReAct; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / ReActCompleted unsafe ratio0.08ratioLong-horizon tasks#S4.T2; row GPT-4 / ReAct; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / ReActIncomplete ratio0.88ratioLong-horizon tasks#S4.T2; row GPT-4 / ReAct; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / PCA-EVALRejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / PCA-EVALRisk ratio (goal evaluator)0.36ratioDetailed hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / PCA-EVALRisk ratio (LLM evaluator)0.17ratioDetailed hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / PCA-EVALExecution ratio0.85ratioDetailed hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / PCA-EVALExecution time97.30sDetailed hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Detailed hazardous tasks; column Execution time
GPT-4 / PCA-EVALRejection ratio0.00ratioAbstract hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / PCA-EVALRisk ratio0.17ratioAbstract hazardous tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / PCA-EVALCompleted safe ratio0.06ratioLong-horizon tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / PCA-EVALCompleted unsafe ratio0.06ratioLong-horizon tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / PCA-EVALIncomplete ratio0.88ratioLong-horizon tasks#S4.T2; row GPT-4 / PCA-EVAL; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / KARMARejection ratio0.00ratioDetailed hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Detailed hazardous tasks; column Rejection ratio
GPT-4 / KARMARisk ratio (goal evaluator)0.47ratioDetailed hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
GPT-4 / KARMARisk ratio (LLM evaluator)0.47ratioDetailed hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
GPT-4 / KARMAExecution ratio0.79ratioDetailed hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Detailed hazardous tasks; column Execution ratio
GPT-4 / KARMAExecution time28.66sDetailed hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Detailed hazardous tasks; column Execution time
GPT-4 / KARMARejection ratio0.00ratioAbstract hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Abstract hazardous tasks; column Rejection ratio
GPT-4 / KARMARisk ratio0.39ratioAbstract hazardous tasks#S4.T2; row GPT-4 / KARMA; condition Abstract hazardous tasks; column Risk ratio
GPT-4 / KARMACompleted safe ratio0.70ratioLong-horizon tasks#S4.T2; row GPT-4 / KARMA; condition Long-horizon tasks; column Completed safe ratio
GPT-4 / KARMACompleted unsafe ratio0.11ratioLong-horizon tasks#S4.T2; row GPT-4 / KARMA; condition Long-horizon tasks; column Completed unsafe ratio
GPT-4 / KARMAIncomplete ratio0.19ratioLong-horizon tasks#S4.T2; row GPT-4 / KARMA; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / Lota-BenchRejection ratio0.11ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / Lota-BenchRisk ratio (goal evaluator)0.54ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / Lota-BenchRisk ratio (LLM evaluator)0.31ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / Lota-BenchExecution ratio0.74ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / Lota-BenchExecution time19.08sDetailed hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Detailed hazardous tasks; column Execution time
Llama3-8B / Lota-BenchRejection ratio0.24ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / Lota-BenchRisk ratio0.32ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / Lota-BenchCompleted safe ratio0.26ratioLong-horizon tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / Lota-BenchCompleted unsafe ratio0.20ratioLong-horizon tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / Lota-BenchIncomplete ratio0.54ratioLong-horizon tasks#A4.T12; row Llama3-8B / Lota-Bench; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / LLM-PlannerRejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / LLM-PlannerRisk ratio (goal evaluator)0.17ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / LLM-PlannerRisk ratio (LLM evaluator)0.13ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / LLM-PlannerExecution ratio0.69ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / LLM-PlannerExecution time54.59sDetailed hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Detailed hazardous tasks; column Execution time
Llama3-8B / LLM-PlannerRejection ratio0.00ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / LLM-PlannerRisk ratio0.15ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / LLM-PlannerCompleted safe ratio0.02ratioLong-horizon tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / LLM-PlannerCompleted unsafe ratio0.02ratioLong-horizon tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / LLM-PlannerIncomplete ratio0.96ratioLong-horizon tasks#A4.T12; row Llama3-8B / LLM-Planner; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / CoELARejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / CoELARisk ratio (goal evaluator)0.03ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / CoELARisk ratio (LLM evaluator)0.01ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / CoELAExecution ratio0.31ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / CoELAExecution time59.54sDetailed hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Detailed hazardous tasks; column Execution time
Llama3-8B / CoELARejection ratio0.00ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / CoELARisk ratio0.02ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / CoELA; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / CoELACompleted safe ratio0.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / CoELA; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / CoELACompleted unsafe ratio0.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / CoELA; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / CoELAIncomplete ratio1.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / CoELA; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / MLDTRejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / MLDTRisk ratio (goal evaluator)0.36ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / MLDTRisk ratio (LLM evaluator)0.40ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / MLDTExecution ratio0.70ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / MLDTExecution time63.56sDetailed hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Detailed hazardous tasks; column Execution time
Llama3-8B / MLDTRejection ratio0.36ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / MLDTRisk ratio0.28ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / MLDT; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / MLDTCompleted safe ratio0.29ratioLong-horizon tasks#A4.T12; row Llama3-8B / MLDT; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / MLDTCompleted unsafe ratio0.22ratioLong-horizon tasks#A4.T12; row Llama3-8B / MLDT; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / MLDTIncomplete ratio0.49ratioLong-horizon tasks#A4.T12; row Llama3-8B / MLDT; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / ProgPromptRejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / ProgPromptRisk ratio (goal evaluator)0.63ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / ProgPromptRisk ratio (LLM evaluator)0.53ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / ProgPromptExecution ratio0.36ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / ProgPromptExecution time37.65sDetailed hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Detailed hazardous tasks; column Execution time
Llama3-8B / ProgPromptRejection ratio0.09ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / ProgPromptRisk ratio0.32ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / ProgPromptCompleted safe ratio0.32ratioLong-horizon tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / ProgPromptCompleted unsafe ratio0.22ratioLong-horizon tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / ProgPromptIncomplete ratio0.46ratioLong-horizon tasks#A4.T12; row Llama3-8B / ProgPrompt; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / ReActRejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / ReActRisk ratio (goal evaluator)0.04ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / ReActRisk ratio (LLM evaluator)0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / ReActExecution ratio0.62ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / ReActExecution time18.08sDetailed hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Detailed hazardous tasks; column Execution time
Llama3-8B / ReActRejection ratio0.00ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / ReActRisk ratio0.06ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / ReAct; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / ReActCompleted safe ratio0.06ratioLong-horizon tasks#A4.T12; row Llama3-8B / ReAct; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / ReActCompleted unsafe ratio0.08ratioLong-horizon tasks#A4.T12; row Llama3-8B / ReAct; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / ReActIncomplete ratio0.86ratioLong-horizon tasks#A4.T12; row Llama3-8B / ReAct; condition Long-horizon tasks; column Incomplete ratio
Llama3-8B / PCA-EVALRejection ratio0.00ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Detailed hazardous tasks; column Rejection ratio
Llama3-8B / PCA-EVALRisk ratio (goal evaluator)0.13ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Llama3-8B / PCA-EVALRisk ratio (LLM evaluator)0.07ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Llama3-8B / PCA-EVALExecution ratio0.50ratioDetailed hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Detailed hazardous tasks; column Execution ratio
Llama3-8B / PCA-EVALExecution time113.26sDetailed hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Detailed hazardous tasks; column Execution time
Llama3-8B / PCA-EVALRejection ratio0.00ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Abstract hazardous tasks; column Rejection ratio
Llama3-8B / PCA-EVALRisk ratio0.04ratioAbstract hazardous tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Abstract hazardous tasks; column Risk ratio
Llama3-8B / PCA-EVALCompleted safe ratio0.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Long-horizon tasks; column Completed safe ratio
Llama3-8B / PCA-EVALCompleted unsafe ratio0.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Long-horizon tasks; column Completed unsafe ratio
Llama3-8B / PCA-EVALIncomplete ratio1.00ratioLong-horizon tasks#A4.T12; row Llama3-8B / PCA-EVAL; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / Lota-BenchRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / Lota-BenchRisk ratio (goal evaluator)0.45ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / Lota-BenchRisk ratio (LLM evaluator)0.32ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / Lota-BenchExecution ratio0.55ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / Lota-BenchExecution time11.09sDetailed hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / Lota-BenchRejection ratio0.05ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / Lota-BenchRisk ratio0.25ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / Lota-BenchCompleted safe ratio0.18ratioLong-horizon tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / Lota-BenchCompleted unsafe ratio0.28ratioLong-horizon tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / Lota-BenchIncomplete ratio0.54ratioLong-horizon tasks#A4.T13; row Qwen2-7B / Lota-Bench; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / LLM-PlannerRisk ratio (goal evaluator)0.25ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / LLM-PlannerRisk ratio (LLM evaluator)0.24ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / LLM-PlannerExecution ratio0.69ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / LLM-PlannerExecution time68.32sDetailed hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / LLM-PlannerRisk ratio0.14ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / LLM-PlannerCompleted safe ratio0.06ratioLong-horizon tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / LLM-PlannerCompleted unsafe ratio0.16ratioLong-horizon tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / LLM-PlannerIncomplete ratio0.78ratioLong-horizon tasks#A4.T13; row Qwen2-7B / LLM-Planner; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / CoELARejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / CoELARisk ratio (goal evaluator)0.11ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / CoELARisk ratio (LLM evaluator)0.02ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / CoELAExecution ratio0.15ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / CoELAExecution time32.59sDetailed hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / CoELARejection ratio0.00ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / CoELARisk ratio0.05ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / CoELA; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / CoELACompleted safe ratio0.04ratioLong-horizon tasks#A4.T13; row Qwen2-7B / CoELA; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / CoELACompleted unsafe ratio0.00ratioLong-horizon tasks#A4.T13; row Qwen2-7B / CoELA; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / CoELAIncomplete ratio0.96ratioLong-horizon tasks#A4.T13; row Qwen2-7B / CoELA; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / MLDTRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / MLDTRisk ratio (goal evaluator)0.33ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / MLDTRisk ratio (LLM evaluator)0.64ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / MLDTExecution ratio0.59ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / MLDTExecution time18.40sDetailed hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / MLDTRejection ratio0.01ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / MLDTRisk ratio0.28ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / MLDT; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / MLDTCompleted safe ratio0.06ratioLong-horizon tasks#A4.T13; row Qwen2-7B / MLDT; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / MLDTCompleted unsafe ratio0.06ratioLong-horizon tasks#A4.T13; row Qwen2-7B / MLDT; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / MLDTIncomplete ratio0.88ratioLong-horizon tasks#A4.T13; row Qwen2-7B / MLDT; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / ProgPromptRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / ProgPromptRisk ratio (goal evaluator)0.38ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / ProgPromptRisk ratio (LLM evaluator)0.45ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / ProgPromptExecution ratio0.32ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / ProgPromptExecution time26.56sDetailed hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / ProgPromptRejection ratio0.10ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / ProgPromptRisk ratio0.28ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / ProgPromptCompleted safe ratio0.15ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / ProgPromptCompleted unsafe ratio0.21ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / ProgPromptIncomplete ratio0.64ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ProgPrompt; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / ReActRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / ReActRisk ratio (goal evaluator)0.12ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / ReActRisk ratio (LLM evaluator)0.02ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / ReActExecution ratio0.01ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / ReActExecution time14.35sDetailed hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / ReActRejection ratio0.00ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / ReActRisk ratio0.05ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / ReAct; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / ReActCompleted safe ratio0.08ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ReAct; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / ReActCompleted unsafe ratio0.04ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ReAct; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / ReActIncomplete ratio0.88ratioLong-horizon tasks#A4.T13; row Qwen2-7B / ReAct; condition Long-horizon tasks; column Incomplete ratio
Qwen2-7B / PCA-EVALRejection ratio0.00ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Detailed hazardous tasks; column Rejection ratio
Qwen2-7B / PCA-EVALRisk ratio (goal evaluator)0.26ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
Qwen2-7B / PCA-EVALRisk ratio (LLM evaluator)0.10ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
Qwen2-7B / PCA-EVALExecution ratio0.58ratioDetailed hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Detailed hazardous tasks; column Execution ratio
Qwen2-7B / PCA-EVALExecution time68.55sDetailed hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Detailed hazardous tasks; column Execution time
Qwen2-7B / PCA-EVALRejection ratio0.00ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Abstract hazardous tasks; column Rejection ratio
Qwen2-7B / PCA-EVALRisk ratio0.08ratioAbstract hazardous tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Abstract hazardous tasks; column Risk ratio
Qwen2-7B / PCA-EVALCompleted safe ratio0.06ratioLong-horizon tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Long-horizon tasks; column Completed safe ratio
Qwen2-7B / PCA-EVALCompleted unsafe ratio0.04ratioLong-horizon tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Long-horizon tasks; column Completed unsafe ratio
Qwen2-7B / PCA-EVALIncomplete ratio0.90ratioLong-horizon tasks#A4.T13; row Qwen2-7B / PCA-EVAL; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / Lota-BenchRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio (goal evaluator)0.66ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / Lota-BenchRisk ratio (LLM evaluator)0.66ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / Lota-BenchExecution ratio0.87ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / Lota-BenchExecution time24.39sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / Lota-BenchRejection ratio0.01ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio0.66ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / Lota-BenchCompleted safe ratio0.46ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / Lota-BenchCompleted unsafe ratio0.32ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / Lota-BenchIncomplete ratio0.22ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Lota-Bench; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio (goal evaluator)0.37ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / LLM-PlannerRisk ratio (LLM evaluator)0.30ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / LLM-PlannerExecution ratio0.80ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / LLM-PlannerExecution time90.50sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio0.34ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / LLM-PlannerCompleted safe ratio0.11ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / LLM-PlannerCompleted unsafe ratio0.11ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / LLM-PlannerIncomplete ratio0.78ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / LLM-Planner; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio (goal evaluator)0.14ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio (LLM evaluator)0.13ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / Co-LLM-AgentExecution ratio0.31ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / Co-LLM-AgentExecution time90.81sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio0.10ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / Co-LLM-AgentCompleted safe ratio0.10ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / Co-LLM-AgentCompleted unsafe ratio0.00ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / Co-LLM-AgentIncomplete ratio0.90ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / Co-LLM-Agent; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / MLDTRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / MLDTRisk ratio (goal evaluator)0.54ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / MLDTRisk ratio (LLM evaluator)0.61ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / MLDTExecution ratio0.78ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / MLDTExecution time26.02sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / MLDTRejection ratio0.01ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / MLDTRisk ratio0.51ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / MLDTCompleted safe ratio0.30ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / MLDTCompleted unsafe ratio0.36ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / MLDTIncomplete ratio0.34ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / MLDT; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / ProgPromptRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRisk ratio (goal evaluator)0.57ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / ProgPromptRisk ratio (LLM evaluator)0.67ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / ProgPromptExecution ratio0.37ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / ProgPromptExecution time19.86sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / ProgPromptRejection ratio0.09ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRisk ratio0.60ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / ProgPromptCompleted safe ratio0.26ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / ProgPromptCompleted unsafe ratio0.28ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / ProgPromptIncomplete ratio0.46ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ProgPrompt; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / ReActRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / ReActRisk ratio (goal evaluator)0.29ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / ReActRisk ratio (LLM evaluator)0.20ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / ReActExecution ratio0.68ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / ReActExecution time32.41sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / ReActRejection ratio0.00ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / ReActRisk ratio0.20ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / ReActCompleted safe ratio0.06ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / ReActCompleted unsafe ratio0.02ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / ReActIncomplete ratio0.92ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / ReAct; condition Long-horizon tasks; column Incomplete ratio
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Detailed hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio (goal evaluator)0.37ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (goal evaluator)
DeepSeek-V2.5 / PCA-EVALRisk ratio (LLM evaluator)0.28ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Detailed hazardous tasks; column Risk ratio (LLM evaluator)
DeepSeek-V2.5 / PCA-EVALExecution ratio0.75ratioDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Detailed hazardous tasks; column Execution ratio
DeepSeek-V2.5 / PCA-EVALExecution time104.13sDetailed hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Detailed hazardous tasks; column Execution time
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract hazardous tasks; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio0.20ratioAbstract hazardous tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract hazardous tasks; column Risk ratio
DeepSeek-V2.5 / PCA-EVALCompleted safe ratio0.14ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Long-horizon tasks; column Completed safe ratio
DeepSeek-V2.5 / PCA-EVALCompleted unsafe ratio0.04ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Long-horizon tasks; column Completed unsafe ratio
DeepSeek-V2.5 / PCA-EVALIncomplete ratio0.82ratioLong-horizon tasks#A4.T14; row DeepSeek-V2.5 / PCA-EVAL; condition Long-horizon tasks; column Incomplete ratio
GPT-4 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L1#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L2#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L3#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L4#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / Lota-BenchRisk ratio0.90ratioAbstract tasks / L1#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L1; column Risk ratio
GPT-4 / Lota-BenchRisk ratio0.54ratioAbstract tasks / L2#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L2; column Risk ratio
GPT-4 / Lota-BenchRisk ratio0.56ratioAbstract tasks / L3#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L3; column Risk ratio
GPT-4 / Lota-BenchRisk ratio0.35ratioAbstract tasks / L4#S5.tab1; row GPT-4 / Lota-Bench; condition Abstract tasks / L4; column Risk ratio
GPT-4 / LLM-PlannerRejection ratio0.10ratioAbstract tasks / L1#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / LLM-PlannerRejection ratio0.12ratioAbstract tasks / L2#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / LLM-PlannerRejection ratio0.40ratioAbstract tasks / L3#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / LLM-PlannerRejection ratio0.63ratioAbstract tasks / L4#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / LLM-PlannerRisk ratio0.48ratioAbstract tasks / L1#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L1; column Risk ratio
GPT-4 / LLM-PlannerRisk ratio0.40ratioAbstract tasks / L2#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L2; column Risk ratio
GPT-4 / LLM-PlannerRisk ratio0.25ratioAbstract tasks / L3#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L3; column Risk ratio
GPT-4 / LLM-PlannerRisk ratio0.15ratioAbstract tasks / L4#S5.tab1; row GPT-4 / LLM-Planner; condition Abstract tasks / L4; column Risk ratio
GPT-4 / CoELARejection ratio0.00ratioAbstract tasks / L1#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / CoELARejection ratio0.00ratioAbstract tasks / L2#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / CoELARejection ratio0.00ratioAbstract tasks / L3#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / CoELARejection ratio0.00ratioAbstract tasks / L4#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / CoELARisk ratio0.35ratioAbstract tasks / L1#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L1; column Risk ratio
GPT-4 / CoELARisk ratio0.18ratioAbstract tasks / L2#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L2; column Risk ratio
GPT-4 / CoELARisk ratio0.08ratioAbstract tasks / L3#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L3; column Risk ratio
GPT-4 / CoELARisk ratio0.25ratioAbstract tasks / L4#S5.tab1; row GPT-4 / CoELA; condition Abstract tasks / L4; column Risk ratio
GPT-4 / MLDTRejection ratio0.05ratioAbstract tasks / L1#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / MLDTRejection ratio0.13ratioAbstract tasks / L2#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / MLDTRejection ratio0.20ratioAbstract tasks / L3#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / MLDTRejection ratio0.20ratioAbstract tasks / L4#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / MLDTRisk ratio0.62ratioAbstract tasks / L1#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L1; column Risk ratio
GPT-4 / MLDTRisk ratio0.31ratioAbstract tasks / L2#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L2; column Risk ratio
GPT-4 / MLDTRisk ratio0.31ratioAbstract tasks / L3#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L3; column Risk ratio
GPT-4 / MLDTRisk ratio0.38ratioAbstract tasks / L4#S5.tab1; row GPT-4 / MLDT; condition Abstract tasks / L4; column Risk ratio
GPT-4 / ProgPromptRejection ratio0.07ratioAbstract tasks / L1#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / ProgPromptRejection ratio0.19ratioAbstract tasks / L2#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / ProgPromptRejection ratio0.28ratioAbstract tasks / L3#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / ProgPromptRejection ratio0.27ratioAbstract tasks / L4#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / ProgPromptRisk ratio0.62ratioAbstract tasks / L1#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L1; column Risk ratio
GPT-4 / ProgPromptRisk ratio0.54ratioAbstract tasks / L2#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L2; column Risk ratio
GPT-4 / ProgPromptRisk ratio0.15ratioAbstract tasks / L3#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L3; column Risk ratio
GPT-4 / ProgPromptRisk ratio0.31ratioAbstract tasks / L4#S5.tab1; row GPT-4 / ProgPrompt; condition Abstract tasks / L4; column Risk ratio
GPT-4 / MAPRejection ratio0.00ratioAbstract tasks / L1#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / MAPRejection ratio0.00ratioAbstract tasks / L2#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / MAPRejection ratio0.00ratioAbstract tasks / L3#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / MAPRejection ratio0.00ratioAbstract tasks / L4#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / MAPRisk ratio0.54ratioAbstract tasks / L1#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L1; column Risk ratio
GPT-4 / MAPRisk ratio0.31ratioAbstract tasks / L2#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L2; column Risk ratio
GPT-4 / MAPRisk ratio0.08ratioAbstract tasks / L3#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L3; column Risk ratio
GPT-4 / MAPRisk ratio0.23ratioAbstract tasks / L4#S5.tab1; row GPT-4 / MAP; condition Abstract tasks / L4; column Risk ratio
GPT-4 / ReActRejection ratio0.19ratioAbstract tasks / L1#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / ReActRejection ratio0.24ratioAbstract tasks / L2#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / ReActRejection ratio0.35ratioAbstract tasks / L3#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / ReActRejection ratio0.48ratioAbstract tasks / L4#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / ReActRisk ratio0.70ratioAbstract tasks / L1#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L1; column Risk ratio
GPT-4 / ReActRisk ratio0.59ratioAbstract tasks / L2#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L2; column Risk ratio
GPT-4 / ReActRisk ratio0.52ratioAbstract tasks / L3#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L3; column Risk ratio
GPT-4 / ReActRisk ratio0.41ratioAbstract tasks / L4#S5.tab1; row GPT-4 / ReAct; condition Abstract tasks / L4; column Risk ratio
GPT-4 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L1#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L2#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L3#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L4#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / PCA-EVALRisk ratio0.27ratioAbstract tasks / L1#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L1; column Risk ratio
GPT-4 / PCA-EVALRisk ratio0.11ratioAbstract tasks / L2#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L2; column Risk ratio
GPT-4 / PCA-EVALRisk ratio0.12ratioAbstract tasks / L3#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L3; column Risk ratio
GPT-4 / PCA-EVALRisk ratio0.20ratioAbstract tasks / L4#S5.tab1; row GPT-4 / PCA-EVAL; condition Abstract tasks / L4; column Risk ratio
GPT-4 / KARMARejection ratio0.00ratioAbstract tasks / L1#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L1; column Rejection ratio
GPT-4 / KARMARejection ratio0.00ratioAbstract tasks / L2#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L2; column Rejection ratio
GPT-4 / KARMARejection ratio0.00ratioAbstract tasks / L3#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L3; column Rejection ratio
GPT-4 / KARMARejection ratio0.00ratioAbstract tasks / L4#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L4; column Rejection ratio
GPT-4 / KARMARisk ratio0.54ratioAbstract tasks / L1#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L1; column Risk ratio
GPT-4 / KARMARisk ratio0.48ratioAbstract tasks / L2#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L2; column Risk ratio
GPT-4 / KARMARisk ratio0.22ratioAbstract tasks / L3#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L3; column Risk ratio
GPT-4 / KARMARisk ratio0.32ratioAbstract tasks / L4#S5.tab1; row GPT-4 / KARMA; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / Lota-BenchRejection ratio0.00ratioAbstract tasks / L1#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / Lota-BenchRejection ratio0.10ratioAbstract tasks / L2#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / Lota-BenchRejection ratio0.31ratioAbstract tasks / L3#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / Lota-BenchRejection ratio0.56ratioAbstract tasks / L4#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / Lota-BenchRisk ratio0.49ratioAbstract tasks / L1#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / Lota-BenchRisk ratio0.43ratioAbstract tasks / L2#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / Lota-BenchRisk ratio0.23ratioAbstract tasks / L3#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / Lota-BenchRisk ratio0.13ratioAbstract tasks / L4#A4.T15; row Llama3-8B / Lota-Bench; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L1#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L3#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L4#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / LLM-PlannerRisk ratio0.11ratioAbstract tasks / L1#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / LLM-PlannerRisk ratio0.14ratioAbstract tasks / L2#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / LLM-PlannerRisk ratio0.14ratioAbstract tasks / L3#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / LLM-PlannerRisk ratio0.20ratioAbstract tasks / L4#A4.T15; row Llama3-8B / LLM-Planner; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L1#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L3#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L4#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / Co-LLM-AgentRisk ratio0.02ratioAbstract tasks / L1#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / Co-LLM-AgentRisk ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / Co-LLM-AgentRisk ratio0.00ratioAbstract tasks / L3#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / Co-LLM-AgentRisk ratio0.06ratioAbstract tasks / L4#A4.T15; row Llama3-8B / Co-LLM-Agent; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / MLDTRejection ratio0.08ratioAbstract tasks / L1#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / MLDTRejection ratio0.17ratioAbstract tasks / L2#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / MLDTRejection ratio0.55ratioAbstract tasks / L3#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / MLDTRejection ratio0.63ratioAbstract tasks / L4#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / MLDTRisk ratio0.44ratioAbstract tasks / L1#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / MLDTRisk ratio0.21ratioAbstract tasks / L2#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / MLDTRisk ratio0.18ratioAbstract tasks / L3#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / MLDTRisk ratio0.31ratioAbstract tasks / L4#A4.T15; row Llama3-8B / MLDT; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / ProgPromptRejection ratio0.03ratioAbstract tasks / L1#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / ProgPromptRejection ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / ProgPromptRejection ratio0.06ratioAbstract tasks / L3#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / ProgPromptRejection ratio0.26ratioAbstract tasks / L4#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / ProgPromptRisk ratio0.46ratioAbstract tasks / L1#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / ProgPromptRisk ratio0.47ratioAbstract tasks / L2#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / ProgPromptRisk ratio0.21ratioAbstract tasks / L3#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / ProgPromptRisk ratio0.15ratioAbstract tasks / L4#A4.T15; row Llama3-8B / ProgPrompt; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / ReActRejection ratio0.00ratioAbstract tasks / L1#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / ReActRejection ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / ReActRejection ratio0.00ratioAbstract tasks / L3#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / ReActRejection ratio0.00ratioAbstract tasks / L4#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / ReActRisk ratio0.07ratioAbstract tasks / L1#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / ReActRisk ratio0.05ratioAbstract tasks / L2#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / ReActRisk ratio0.06ratioAbstract tasks / L3#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / ReActRisk ratio0.04ratioAbstract tasks / L4#A4.T15; row Llama3-8B / ReAct; condition Abstract tasks / L4; column Risk ratio
Llama3-8B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L1#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L1; column Rejection ratio
Llama3-8B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L2#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L2; column Rejection ratio
Llama3-8B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L3#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L3; column Rejection ratio
Llama3-8B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L4#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L4; column Rejection ratio
Llama3-8B / PCA-EVALRisk ratio0.06ratioAbstract tasks / L1#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L1; column Risk ratio
Llama3-8B / PCA-EVALRisk ratio0.02ratioAbstract tasks / L2#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L2; column Risk ratio
Llama3-8B / PCA-EVALRisk ratio0.03ratioAbstract tasks / L3#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L3; column Risk ratio
Llama3-8B / PCA-EVALRisk ratio0.04ratioAbstract tasks / L4#A4.T15; row Llama3-8B / PCA-EVAL; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / Lota-BenchRejection ratio0.00ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / Lota-BenchRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / Lota-BenchRejection ratio0.03ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / Lota-BenchRejection ratio0.16ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / Lota-BenchRisk ratio0.29ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / Lota-BenchRisk ratio0.32ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / Lota-BenchRisk ratio0.19ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / Lota-BenchRisk ratio0.21ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / Lota-Bench; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / LLM-PlannerRisk ratio0.18ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / LLM-PlannerRisk ratio0.22ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / LLM-PlannerRisk ratio0.03ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / LLM-PlannerRisk ratio0.13ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / LLM-Planner; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / Co-LLM-AgentRisk ratio0.06ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / Co-LLM-AgentRisk ratio0.06ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / Co-LLM-AgentRisk ratio0.04ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / Co-LLM-AgentRisk ratio0.04ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / Co-LLM-Agent; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / MLDTRejection ratio0.03ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / MLDTRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / MLDTRejection ratio0.02ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / MLDTRejection ratio0.00ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / MLDTRisk ratio0.25ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / MLDTRisk ratio0.55ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / MLDTRisk ratio0.20ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / MLDTRisk ratio0.10ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / MLDT; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / ProgPromptRejection ratio0.07ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / ProgPromptRejection ratio0.08ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / ProgPromptRejection ratio0.09ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / ProgPromptRejection ratio0.15ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / ProgPromptRisk ratio0.42ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / ProgPromptRisk ratio0.26ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / ProgPromptRisk ratio0.20ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / ProgPromptRisk ratio0.25ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / ProgPrompt; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / ReActRejection ratio0.00ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / ReActRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / ReActRejection ratio0.00ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / ReActRejection ratio0.00ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / ReActRisk ratio0.02ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / ReActRisk ratio0.05ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / ReActRisk ratio0.08ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / ReActRisk ratio0.04ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / ReAct; condition Abstract tasks / L4; column Risk ratio
Qwen2-7B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L1; column Rejection ratio
Qwen2-7B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L2; column Rejection ratio
Qwen2-7B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L3; column Rejection ratio
Qwen2-7B / PCA-EVALRejection ratio0.00ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L4; column Rejection ratio
Qwen2-7B / PCA-EVALRisk ratio0.12ratioAbstract tasks / L1#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L1; column Risk ratio
Qwen2-7B / PCA-EVALRisk ratio0.05ratioAbstract tasks / L2#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L2; column Risk ratio
Qwen2-7B / PCA-EVALRisk ratio0.04ratioAbstract tasks / L3#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L3; column Risk ratio
Qwen2-7B / PCA-EVALRisk ratio0.12ratioAbstract tasks / L4#A4.T16; row Qwen2-7B / PCA-EVAL; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRejection ratio0.00ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRejection ratio0.02ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio0.82ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio0.58ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio0.56ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / Lota-BenchRisk ratio0.66ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / Lota-Bench; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRejection ratio0.00ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio0.48ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio0.35ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio0.32ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / LLM-PlannerRisk ratio0.19ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / LLM-Planner; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRejection ratio0.00ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio0.22ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio0.08ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio0.04ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / Co-LLM-AgentRisk ratio0.04ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / Co-LLM-Agent; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / MLDTRejection ratio0.03ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / MLDTRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / MLDTRejection ratio0.02ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / MLDTRejection ratio0.00ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / MLDTRisk ratio0.70ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / MLDTRisk ratio0.50ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / MLDTRisk ratio0.40ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / MLDTRisk ratio0.45ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / MLDT; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / ProgPromptRejection ratio0.03ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRejection ratio0.07ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRejection ratio0.13ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRejection ratio0.13ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / ProgPromptRisk ratio0.75ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / ProgPromptRisk ratio0.65ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / ProgPromptRisk ratio0.55ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / ProgPromptRisk ratio0.45ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / ProgPrompt; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / ReActRejection ratio0.00ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / ReActRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / ReActRejection ratio0.00ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / ReActRejection ratio0.00ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / ReActRisk ratio0.24ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / ReActRisk ratio0.18ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / ReActRisk ratio0.22ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / ReActRisk ratio0.17ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / ReAct; condition Abstract tasks / L4; column Risk ratio
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L1; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L2; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L3; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRejection ratio0.00ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L4; column Rejection ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio0.28ratioAbstract tasks / L1#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L1; column Risk ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio0.11ratioAbstract tasks / L2#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L2; column Risk ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio0.22ratioAbstract tasks / L3#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L3; column Risk ratio
DeepSeek-V2.5 / PCA-EVALRisk ratio0.17ratioAbstract tasks / L4#A4.T17; row DeepSeek-V2.5 / PCA-EVAL; condition Abstract tasks / L4; column Risk ratio
Reading scope & review

Selected original HTML sections and table context; not a complete paper/appendix audit or independent reproduction.

Model-assisted reading; independent evidence review pending.

Check the original source ↗