Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States

Researchers introduced Mind2Dialogue, a method for training human-aware language models by simulating user mental states, to address what the work calls a fundamental supervision gap in current datasets for LLM assistant training. The approach targets long-term collaboration in learning, reasoning, and decision-making, where the authors argue models need a deeper understanding of the people they serve.

As language models become more capable, long-term collaboration in learning, reasoning, and decision-making calls for a deeper understanding of the people they serve. Yet training such human-aware language models faces a fundamental supervision gap because current datasets for LLM assistant training