List your company

Causal Discovery on CausalMan Small

48.3F1 ScoreCAST (HITL)
-1.72411.26324.2537.237Jul 2, 2026

Evaluation results

MethodLinks
CAST (HITL)LLM=Qwen 3 14B2026.0748.360.840.2102.41,856.65
CASTLLM Backbone=Qw.3 14B, HITL=true, RAG=false2026.0748.360.8—102.4—
CAST (RAG)LLM=Qwen 3 14B2026.0742.665.633.5111.4944.92
CASTLLM Backbone=Qw.3 14B, HITL=false, RAG=true2026.0742.665.6—111.4—
CAST (HITL+RAG)LLM=Qwen 3 14B2026.0742.472.430.71031,064.03
CASTLLM Backbone=Qw.3 14B, HITL=true, RAG=true2026.0742.472.4—103—
CAST (HITL)LLM=o3-mini2026.0730.448.522.4112.2254.66
CausalStewardLLM Backbone=o3-mini, HITL=true, RAG=false2026.0730.448.5—112.2—
CAST (HITL+RAG)LLM=o3-mini2026.0730.345.422.8118.4316.65
CausalStewardLLM Backbone=o3-mini, HITL=true, RAG=true2026.0730.345.4—118.4—
LLM-BFSLLM=o3-mini2026.0728.3100—315.2—
LLM-BFSLLM Backbone=o3-mini2026.0728.3100—315.2—
CAST (RAG)LLM=o3-mini2026.0723.537.317.8121.4455.31
CausalStewardLLM Backbone=o3-mini, HITL=false, RAG=true2026.0723.537.3—121.4—
PairwiseLLM=Qwen3-14B2026.0715.39.3—583.5—
PairwiseLLM Backbone=Qwen3-14B2026.0715.39.3—583.5—
CAST (HITL+RAG)LLM=gpt 4o-mini2026.0714.221.710.5140.41,630.19
CASTLLM Backbone=o4-mini, HITL=true, RAG=true2026.0714.221.7—140.4—
CAST (HITL)LLM=gpt 4o-mini2026.0713.932.69.6126.81,456.33
CASTLLM Backbone=o4-mini, HITL=true, RAG=false2026.0713.932.6—126.8—
PairwiseLLM=o3-mini2026.0712.67.442.1710—
CAST (RAG)LLM=gpt 4o-mini2026.0711.316.98.9145913.15
CASTLLM Backbone=o4-mini, HITL=false, RAG=true2026.0711.316.9—145—
CausalCopilot2026.077.8264.8117.4—
CausalCopilot2026.077.826—117.4—
XGES2026.075.63.711.4448.825,173.8
XGES2026.075.63.7—448.8—
BOSS2026.075.42.9—864.2—
BOSS2026.075.42.9—864.2—
FCI2026.072.82.14.3221.49.37
FCI2026.072.82.1—221.4—
GranDAG2026.070.21.80.2113.4333.43
GranDAG2026.070.21.8—113.4—