OmniToM - Belief Labeling - Context: leaderboard

OmniToM (2026) evaluates theory of mind through explicit belief modeling on 895 short social stories from ToMBench, with gold belief propositions built by a human-calibrated annotation pipeline. This board is Stage 2 labeling accuracy for the context (deceptive, temporal, counterfactual or neutral) dimension of the given belief propositions; higher is better.

Metric: Belief-labeling accuracy (%): exact-match accuracy of the predicted schema label on the given gold belief propositions, macro-averaged over stories, for the context (deceptive, temporal, counterfactual or neutral) dimension, on the 895 OmniToM stories from seven ToMBench categories (22,343 gold belief propositions), zero-shot TELeR level 3 prompts; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Mistral Large 2 (Jul)92.97#1144
2Gemini 2.5 Flash92.14#463
3Gemma 3 27B (IT)92.07#1324
4Mistral Small 391.9#1796
5Llama 3.3 70B Instruct91.69#1385
6Qwen 3 32B90.81#972
7GPT-588.83#178
8Llama 3.1 8B Instruct76.81#4600
9Qwen 3 8B74.62#1873

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=omnitom-belief-labeling-context · How It Works · Data refreshed daily, snapshot 2026-11-21.