New CToM framework trains LLMs to anticipate user responses for better collaboration

Researchers have developed a new framework called Collaborative Theory of Mind (CToM) designed to improve how large language models (LLMs) collaborate with users. CToM trains LLMs to anticipate user responses and use these predictions to refine their own behavior, leading to better task performance and efficiency. Experiments across mathematical reasoning, coding, and writing tasks showed CToM outperforming baseline models, notably increasing accuracy in collaborative mathematical reasoning and reducing the number of user interactions required for coding and writing tasks. AI

IMPACT Enhances LLM collaboration by enabling models to predict and adapt to user input, potentially improving user experience and task completion rates.

RANK_REASON The cluster contains an academic paper detailing a new framework for LLM collaboration. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Collaborative Theory of Mind
• CToM
• Group Relative Policy Optimization
• GRPO
• Hugging Face
• Large language models
• LLMs

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →