AI Paper Atlas

Research map / Large language models / Agents, code and math

ToMAP: Training Opponent-Aware LLM Persuaders with Theory of Mind

Peixuan Han, Zijia Liu, Jiaxuan You

ICML 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add RL trainingQwen2.5-3B · Table 1 avgagreement shift: 11.76 -> 14.60Improved
add SFT on successful conversationsQwen2.5-3B · Table 1 avgagreement shift: 11.76 -> 12.49No clear effect
add ToM modules on top of RLQwen2.5-3B · Table 1 avgagreement shift: 14.60 -> 17.48Improved
add ToM modules on top of RLLLaMa3.2-3B · Table 1 avgagreement shift: 21.76 -> 24.35Improved
add ToM info to untrained base model (ablation)Qwen2.5-3B · Qwen2.5-7B persuadee, 3 datasetsagreement shift: 8.44 -> 8.18No clear effect
add ToM info zero-shot to larger models (ablation)Gemma-3-27B · Qwen2.5-7B persuadee, 3 datasetsagreement shift: 11.92 -> 13.26Improved
add ToM info to RL-trained model (ablation)Qwen2.5-3B+ToMAP · Qwen2.5-7B persuadee, 3 datasetsagreement shift: 14.47 -> 19.19Improved
add long-CoT prompt (ablation)Qwen2.5-3B · CMVagreement shift: 12.75 -> 14.25No clear effect
add attitude predictor to counterclaim predictor (ablation)ToMAP · CMVagreement shift: 15.57 -> 23.01Improved
add attitude predictor to counterclaim predictor (ablation)ToMAP · Anthropicagreement shift: 9.55 -> 15.82Improved
use ground-truth attitudes instead of predicted (ablation)ToMAP · CMVagreement shift: 23.01 -> 23.57No clear effect
add ToM modules to RLQwen2.5-3B · LLaMa3.1-8B persuadee, CMV (OOD)agreement shift: 12.15 -> 10.77Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy