MuSeToM - Task 1 - F1: leaderboard

MuSeToM - Task 1 - 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
1MiniMax-M2.586.05#584
2Qwen 3.5 397B A17B85.34#251
3QwQ-32B82.61#915
4DeepSeek V3.182.57#502
5GPT-OSS-120B82.53#703
6MiniMax-M2.178.54#574
7DeepSeek R177.54#478
8Llama 3.1 405B Instruct77.25#1122
9DeepSeek V3.275.57#412
10GPT-4o70.09#744
11Qwen 3 235B A22B67.75#628
12Llama 4 Maverick Instruct67.51#1163
13Kimi K2 (Thinking)65.5
14Mixtral 8x22B Instruct63.35#2097
15Phi-3-medium-4k-instruct55.72#2696

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-1-f1 · How It Works · Data refreshed daily, snapshot 2026-11-16.