List your company

Theory of Mind reasoning on Hi-ToM

100Accuracy (0th-Order)CoT
97.498.07598.7599.425Jun 10, 2026

Evaluation results

MethodLinks
CoTModel=GPT-5.42026.0610088.756068.7571.2577.75
SIMTOMModel=GPT-5.42026.0610092.588.7587.58089.75
TIMETOMModel=GPT-5.42026.0610092.59087.581.2590.25
RECTOMModel=GPT-5.42026.06100100100100100100
CoTModel=Gemini-32026.0610010092.58583.7592.25
SIMTOMModel=Gemini-32026.0610091.259088.7586.2591.25
TIMETOMModel=Gemini-32026.0610078.7586.2586.2587.587.75
RECTOMModel=Gemini-32026.0610010097.597.597.598.5
CoTModel=Qwen3.52026.061009588.7583.758590.5
SIMTOMModel=Qwen3.52026.0610087.5909086.2590.75
TIMETOMModel=Qwen3.52026.061007087.586.2582.585.25
RECTOMModel=Qwen3.52026.06100100100100100100
CoTModel=Gemma-42026.0610093.756562.558.7576
RECTOMModel=Gemma-42026.0610010010010097.599.5
SIMTOMModel=Gemma-42026.0698.7591.2581.2573.7573.7583.75
TIMETOMModel=Gemma-42026.0697.5706063.756070.25