DialToM - Prospective (Persuasion): leaderboard
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 January 2027. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 86.96 |
| 2 | Gemini 2.5 Pro | 19.57 |
| 3 | Gemini 2.5 Flash | 17.93 |
| 4 | Qwen 3 235B A22B | 17.93 |
| 5 | GPT-5 | 16.85 |
| 6 | Kimi K2 | 13.04 |
| 7 | GPT-4.1 | 10.87 |
| 8 | Llama 3.1 8B | 10.33 |
| 9 | Mistral Small 3.2 | 7.6 |
| 10 | Llama 3.3 70B | 7.07 |
| 11 | GPT-OSS-120B | 6.52 |
| 12 | Llama 4 Maverick | 5.6 |
| 13 | Mistral Nemo | 5.44 |
Interactive version: theaggregate.ai/benchmark?slug=dialtom-prospective-persuasion · How It Works · Data refreshed daily, snapshot 2026-10-07.