MuSeToM - Task 3 - F1: leaderboard

MuSeToM - Task 3 - F1 — MuSeToM v1 Table1 direct prompting on human-validated textual belief revision; clean tasks1binary/2multiplechoice and noisy tasks3binary/4multiplechoice remain separate. Accuracy and Macro-F1 are distinct percentage metrics. AppendixB.1 binds the Llama/Mistral/Mixtral instruction checkpoints; Tables23and28–31bind the sole Phi row to Phi-3-Medium-4K-Instruct in the same18-model cohort. temperature0,top_p1,1024output tokens for the listed first group and10240for later reasoning models; Phi output budget is unreported because B.1omits it. Effort/API dated snapshots are unreported. Eighteen model rows; Human reference is a separate reference, not Median Human/Expert and not an AI model. Main table shorthand is resolved from this paper only; no global Base/Instruct or generation alias. Metric: Macro-F1 (%).

Metric: Macro-F1 (%). Source: arxiv.org. Saturation forecast: Around February 2027. 18 models tracked. Updated 15 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-OSS-120B80.48#703
2MiniMax-M2.579.83#584
3QwQ-32B77.07#915
4Kimi K2 (Thinking)77.05
5MiniMax-M2.176.67#574
6GPT-4o71.72#744
7Llama 3.1 405B Instruct71.13#1122
8DeepSeek V3.269.86#412
9Qwen 3 235B A22B68.71#628
10DeepSeek R168.24#478
11Qwen 3.5 397B A17B67.81#251
12Mixtral 8x22B Instruct65.67#2097
13GPT-3.5 Turbo64.51#2973
14Phi-3-medium-4k-instruct63.37#2696
15Llama 4 Maverick Instruct62.35#1163

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=musetom-task-3-f1 · How It Works · Data refreshed daily, snapshot 2026-11-16.