ToM2Vec: Continuous Theory-of-Mind Tokens for Multi-Agent Coordination

ToM2Vec learns abstract theory-of-mind representations for partner agents using a fixed budget of continuous latent reasoning tokens. During training, these tokens are supervised by the partner's actual thoughts. At test time, each agent generates its own ToM tokens to condition its thoughts and actions. Evaluated across collaborative, competitive, and mixed-motive environments (Overcooked-AI, multiparty negotiation, Decrypto, Hanabi, Collab Escape/Capture), it reportedly improves over the base model and several baselines. The authors report that performance drops when reasoning is expressed in text, and that training generalizes to partner model families unseen at training time, including GPT-5.4.

Why it matters

Suggests that latent, continuous representations may capture partner beliefs better than text descriptions for LLM agents coordinating under uncertainty.

Read with care

Source claims not independently checked; the source text is an abstract-level summary without full experimental details.