Flipping the Dialogue: Training and Evaluating User Language Models
Conversations with LMs involve two participants: a human user leading the conversation, and an LM assistant responding to the user’s request. To satisfy this specific role, LMs are post-trained to be helpful assistants – optimized to produce exhaustive and well-structured responses, free of ambiguity and grammar errors. User utterances, on the other hand, are rarely perfected, with each user phrasing requests in unique ways, sometimes putting in partial effort at each turn and refining on the fly. To evaluate LM performance in realistic settings, prior work simulated users in multi-turn conversations, often prompting an LLM originally trained to be a helpful assistant to act as a user. However, we show that assistant LMs make for poor user simulators, with the surprising finding that better assistants yield worse simulators. Instead, we introduce purpose-built User Language Models (User LMs) - models post-trained to simulate human users in multi-turn conversations. Through various eval
Flipping the Dialogue: Training and Evaluating User Language Models Tarek Naous ††thanks: Work done while interning at Microsoft Research. Affiliation: Microsoft Research Affiliation: Georgia Institute of Technologytareknaous@gatech.edu ; plaban@microsoft.com Wei Xu Affiliation: Georgia Institute of Technologytareknaous@gatech.edu ; plaban@microsoft.com Jennifer Neville Affiliation: Microsoft Research Abstract Conversations with LMs involve two participants: a human user leading the conversation, and an LM assistant responding to the user’s request. To satisfy this specific role,…
saved by
related reading
- HumanLM: Simulating Users with State Alignment Beats Response Imitationarxiv.org
- Quantifying the Utility of User Simulators for Building Collaborative LLM Assistantsarxiv.org
- Language Models, World Models, and Human Model-Buildinglingo.csail.mit.edu
- [2203.02155] Training language models to follow instructions with human feedbackarxiv.org
- Training language models to follow instructions with human feedback.pdfproceedings.neurips.cc
- user_interactions.pdfself-distillation.github.io
- [2508.19227] Generative Interfaces for Language Modelsarxiv.org
- Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluationsarxiv.org
- [2606.06614] Re-Centering Humans in LLM Personalizationarxiv.org
- Language Models can Solve Computer Tasksarxiv.org
- LLMs Get Lost in Evolving User Intentarxiv.org
- [2511.08579] Training Language Models to Explain Their Own Computationsarxiv.org