ToM2Vec: continuous theory-of-mind tokens improve multi-agent coordination
UNC-ChapelHill · hf · 2026-10-22
Researchers from UNC Chapel Hill, Cisco, and UT Austin introduce ToM2Vec, a method that forms abstract partner-state representations in a rich latent space using a fixed budget of continuous reasoning tokens. Targeting the weakness of LLM agents in reasoning under uncertainty about multiple possible partner states, ToM2Vec improves multi-agent collaboration and coordination across diverse settings. Paper updated Oct 21 and available on Hugging Face.
Related event: ToM2Vec: Continuous Theory-of-Mind vectors help LLM agents model teammates(5 posts)→
More from Research
- Sheffield advertises funded PhD on reliability and bias of LLM-based evaluation — nikaletras · 2026-10-22
- Scientific bottlenecks: reinvent assays, don't just automate them — owl_posting · 2026-10-22
- AI academic writing's vague-language trap: it reads as consistent to you but confuses everyone else — anshulkundaje · 2026-10-22
- YC hosts AI for Biology Paper Club as 81% of US doctors adopt AI — ycombinator · 2026-10-22
- CONCERT: Generative Modeling of Spatially Resolved Perturbation Responses — anshulkundaje · 2026-10-22
- ToM2Vec trained via self-play with Qwen3-8B transfers to GPT-5.4 teams, adding 17.3 points — EliasEskin · 2026-10-22