OmniToM - Belief Extraction - Ambiguous Story Task: 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 1 belief extraction on the 98 Ambiguous Story Task stories, macro F1 against the gold belief propositions; higher is better.

Metric: Belief-extraction macro F1 (%): story-level F1 between the extracted world-fact and actor-belief propositions and the gold propositions, matched semantically by a GPT-5 judge, macro-averaged over stories, on the 98 Ambiguous Story Task stories, on the 895 OmniToM stories from seven ToMBench categories (22,343 gold belief propositions), zero-shot TELeR level 3 prompts; GPT-5 is not scored as it is the judge; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 8 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Mistral Small 352.97#1809
2Gemma 3 27B (IT)48.72#1341
3Mistral Large 2 (Jul)47.75#1164
4Qwen 3 32B46.88#983
5Gemini 2.5 Flash42.4#466
6Qwen 3 8B39.12#1890
7Llama 3.3 70B Instruct37.51#1394
8Llama 3.1 8B Instruct26.34#4655

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=omnitom-belief-extraction-ambiguous-story-task · How It Works · Data refreshed daily, snapshot 2026-11-22.