DialToM - Prospective (Persuasion): leaderboard

Prospective theory of mind on DialToM (2026), naturalistic human-human dialogues from persuasion dialogues (PersuasionForGood): the model must forecast which next dialogue trajectory fits a given mental-state profile, without the dialogue context, choosing among four annotator-verified options. Correctness rate in percent; higher is better.

Metric: Correctness rate (%) on four-option multiple-choice questions of the DialToM Prospective task: forecast the state-consistent next dialogue trajectory from an isolated mental-state profile with no dialogue context (the State-Driven Diagnostic Probe), persuasion domain (naturalistic human-human dialogues from PersuasionForGood; GPT-4o-drafted options verified by professional annotators and Dawid-Skene aggregation), zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro (Preview)86.96#119
2Gemini 2.5 Pro19.57#276
3Gemini 2.5 Flash17.93#463
4Qwen 3 235B A22B17.93#628
5GPT-516.85#178
6Kimi K213.04#443
7GPT-4.110.87#505
8Llama 3.1 8B10.33#5196
9DialToM DeepSeek v3 (checkpoint unspecified)10.33
10Mistral Small 3.27.6#1641
11Llama 3.3 70B7.07#1514
12GPT-OSS-120B6.52#704
13Llama 4 Maverick5.6#1110
14Mistral Nemo5.44#4484

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

Interactive version: theaggregate.ai/benchmark?slug=dialtom-prospective-persuasion · How It Works · Data refreshed daily, snapshot 2026-11-21.