# Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language Models

arXiv: https://arxiv.org/abs/2604.02315 · HTML: https://mlsanity.com/paper/2604.02315/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 38 references that are in the map.
- First posted on arXiv 2 Apr 2026.

## Builds on (most cited first, 30 of 31)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models](https://mlsanity.com/paper/2508.06471/) (2025)
- [HealthBench: Evaluating Large Language Models Towards Improved Human Health](https://mlsanity.com/paper/2505.08775/) (2025)
- [ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL](https://mlsanity.com/paper/2402.19446/) (2024)
- [Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs](https://mlsanity.com/paper/2210.13312/) (2022)
- [Generalizing Verifiable Instruction Following](https://mlsanity.com/paper/2507.02833/) (2025)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions](https://mlsanity.com/paper/2310.15421/) (2023)
- [CollabLLM: From Passive Responders to Active Collaborators](https://mlsanity.com/paper/2502.00640/) (2025)
- [STaR-GATE: Teaching Language Models to Ask Clarifying Questions](https://mlsanity.com/paper/2403.19154/) (2024)
- [OpenClaw-RL: Train Any Agent Simply by Talking](https://mlsanity.com/paper/2603.10165/) (2026)
- [A Survey on Proactive Dialogue Systems: Problems, Methods, and Prospects](https://mlsanity.com/paper/2305.02750/) (2023)
- [Flipping the Dialogue: Training and Evaluating User Language Models](https://mlsanity.com/paper/2510.06552/) (2025)
- [SimpleToM: Exposing the Gap between Explicit ToM Inference and Implicit ToM Application in LLMs](https://mlsanity.com/paper/2410.13648/) (2024)
- [Mind the Sim2Real Gap in User Simulation for Agentic Tasks](https://mlsanity.com/paper/2603.11245/) (2026)
- [SimulatorArena: Are User Simulators Reliable Proxies for Multi-Turn Evaluation of AI Assistants?](https://mlsanity.com/paper/2510.05444/) (2025)
- [CRMArena-Pro: Holistic Assessment of LLM Agents Across Diverse Business Scenarios and Interactions](https://mlsanity.com/paper/2505.18878/) (2025)
- [How to Build User Simulators to Train RL-based Dialog Systems](https://mlsanity.com/paper/1909.01388/) (2019)
- [Approximating Interactive Human Evaluation with Self-Play for Open-Domain Dialog Systems](https://mlsanity.com/paper/1906.09308/) (2019)
- [Towards A Holistic Landscape of Situated Theory of Mind in Large Language Models](https://mlsanity.com/paper/2310.19619/) (2023)
- [Aligning Language Models from User Interactions](https://mlsanity.com/paper/2603.12273/) (2026)
- [Echoing: Identity Failures when LLM Agents Talk to Each Other](https://mlsanity.com/paper/2511.09710/) (2025)

## Abstract (arXiv)

Standard LLM benchmarks evaluate the assistant turn: the model generates a response to an input, a verifier scores correctness, and the analysis ends. This paradigm leaves unmeasured whether the LLM encodes any awareness of what follows the assistant response. We propose user-turn generation as a probe of this gap: given a conversation context of user query and assistant response, we let a model generate under the user role. If the model's weights encode interaction awareness, the generated user turn will be a grounded follow-up that reacts to the preceding context. Through experiments across 11 open-weight LLMs (Qwen3.5, gpt-oss, GLM) and 5 datasets (math reasoning, instruction following, conversation), we show that interaction awareness is decoupled from task accuracy. In particular, within the Qwen3.5 family, GSM8K accuracy scales from 41% (0.8B) to 96.8% (397B-A17B), yet genuine follow-up rates under deterministic generation remain near zero. In contrast, higher temperature sampling reveals interaction awareness is latent with follow up rates reaching 22%. Controlled perturbations validate that the proposed probe measures a real property of the model, and collaboration-oriented post-training on Qwen3.5-2B demonstrates an increase in follow-up rates. Our results show that user-turn generation captures a dimension of LLM behavior, interaction awareness, that is unexplored and invisible with current assistant-only benchmarks.