List your company

Causal Reasoning on CaLM

73.9AccuracyUNICO
45.19652.64860.167.552May 24, 2026

Evaluation results

MethodLinks
UNICOBase Model=Qwen3-8B, Data=UNICO2026.0573.9
UNICOBase Model=Qwen3-4B, Data=UNICO2026.0570.3
Olmo3.1-32B-InstructBase Model=Olmo3.1-32B-Instruct2026.0570.1
Qwen3-32BBase Model=Qwen3-32B2026.0569.2
CauGymBase Model=Qwen3-8B, Data=CauGym2026.0567.1
CauGymBase Model=Qwen3-4B, Data=CauGym2026.0565.5
UNICOBase Model=Olmo3-7B-Instruct, Data=UNICO2026.0561.2
CauGymBase Model=Olmo3-7B-Instruct, Data=CauGym2026.0560.6
OriginalBase Model=Qwen3-8B, Data=Original2026.0560.4
OriginalBase Model=Qwen3-4B, Data=Original2026.0559.3
OriginalBase Model=Olmo3-7B-Instruct, Data=Original2026.0556.6
CDCRBase Model=Olmo3-7B-Instruct, Data=CDCR2026.0554.6
CDCRBase Model=Qwen3-8B, Data=CDCR2026.0551.6
CDCRBase Model=Qwen3-4B, Data=CDCR2026.0546.3