List your company

Robotic Planning on LEMMA Single-Agent Underspecified

99Success Rate (SR)SG-CoT
41.856.6571.586.35Mar 18, 2026

Evaluation results

MethodLinks
SG-CoTLLM=Gemini-2.5-Flash2026.039998
InnerMonoLLM=Gemini-2.5-Flash2026.039593
SG-CoTLLM=Gemini-2.5-Flash, Scene Graph Edges=false2026.038754
SG-CoTLLM=Gemini-2.5-Flash, Iterations=12026.038275
ProgPromptLLM=Gemini-2.5-Flash2026.037977
SG-CoTLLM=Qwen3-VL-2B2026.037767
InnerMonoLLM=Qwen3-VL-2B2026.037562
SG-CoTLLM=Qwen3-VL-2B, Scene Graph Edges=false2026.037552
CLARALLM=Gemini-2.5-Flash2026.037060
SG-CoTLLM=Qwen3-VL-2B, Iterations=12026.036962
SG-CoTLLM=Gemini-2.5-Flash, Iterations=02026.036852
CLARALLM=Qwen3-VL-2B2026.036352
ProgPromptLLM=Qwen3-VL-2B2026.035847
SG-CoTLLM=Qwen3-VL-2B, Iterations=02026.034428