DialToM - Retrospective (Motivational Interviewing): leaderboard

Retrospective theory of mind on DialToM (2026), naturalistic human-human dialogues from motivational interviewing sessions (AnnoMI): the model must identify the mental state (belief, desire, intention, emotion, knowledge or trust) of a speaker from the dialogue so far, 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 Retrospective task: identify a speaker's mental state (belief, desire, intention, emotion, knowledge, trust) from the dialogue context, motivational interviewing domain (naturalistic human-human dialogues from AnnoMI; GPT-4o-drafted options verified by professional annotators and Dawid-Skene aggregation), zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 14 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro (Preview)97.71#123
2DialToM DeepSeek v3 (checkpoint unspecified)96.73
3Qwen 3 235B A22B95.1#633
4Gemini 2.5 Flash94.77#466
5GPT-593.46#185
6GPT-4.193.14#506
7Gemini 2.5 Pro92.48#280
8Kimi K292.16#448
9Llama 4 Maverick87.58#1113
10Llama 3.3 70B77.45#1523
11GPT-OSS-120B75.49#712
12Mistral Small 3.261.11#1647
13Mistral Nemo42.16#4465
14Llama 3.1 8B30.39#5193

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

Interactive version: theaggregate.ai/benchmark?slug=dialtom-retrospective-motivational-interviewing · How It Works · Data refreshed daily, snapshot 2026-11-22.