# Teaching Language Models To Gather Information Proactively

arXiv: https://arxiv.org/abs/2507.21389 · HTML: https://mlsanity.com/paper/2507.21389/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 30 references that are in the map.
- First posted on arXiv 28 Jul 2025.

## Builds on (most cited first, 21 of 21)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [Improve Mathematical Reasoning in Language Models by Automated Process Supervision](https://mlsanity.com/paper/2406.06592/) (2024)
- [QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization](https://mlsanity.com/paper/2104.05938/) (2021)
- [AlphaMath Almost Zero: Process Supervision without Process](https://mlsanity.com/paper/2405.03553/) (2024)
- [CollabLLM: From Passive Responders to Active Collaborators](https://mlsanity.com/paper/2502.00640/) (2025)
- [Are Large Language Models Capable of Generating Human-Level Narratives?](https://mlsanity.com/paper/2407.13248/) (2024)
- [Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training](https://mlsanity.com/paper/2406.00222/) (2024)
- [PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance](https://mlsanity.com/paper/2210.08817/) (2022)
- [Learning to Ask Appropriate Questions in Conversational Recommendation](https://mlsanity.com/paper/2105.04774/) (2021)
- [I Cast Detect Thoughts: Learning to Converse and Guide with Intents and Theory-of-Mind in Dungeons and Dragons](https://mlsanity.com/paper/2212.10060/) (2022)
- [Planning and Editing What You Retrieve for Enhanced Tool Learning](https://mlsanity.com/paper/2404.00450/) (2024)
- [Asking Clarifying Questions Based on Negative Feedback in Conversational Search](https://mlsanity.com/paper/2107.05760/) (2021)
- [Empowering Language Models with Active Inquiry for Deeper Understanding](https://mlsanity.com/paper/2402.03719/) (2024)
- [DOLOMITES: Domain-Specific Long-Form Methodical Tasks](https://mlsanity.com/paper/2405.05938/) (2024)
- [Factored Agents: Decoupling In-Context Learning and Memorization for Robust Tool Use](https://mlsanity.com/paper/2503.22931/) (2025)

## Built on by (most cited first, 14 of 14)

- [HiL-Bench (Human-in-Loop Benchmark): Do Agents Know When to Ask for Help?](https://mlsanity.com/paper/2604.09408/) (2026)
- [Optimizing Diversity and Quality through Base-Aligned Model Collaboration](https://mlsanity.com/paper/2511.05650/) (2025)
- [InfoGatherer: Principled Information Seeking via Evidence Retrieval and Strategic Questioning](https://mlsanity.com/paper/2603.05909/) (2026)
- [IntentRL: Training Proactive User-intent Agents for Open-ended Deep Research via Reinforcement Learning](https://mlsanity.com/paper/2602.03468/) (2026)
- [SCICONVBENCH: Benchmarking LLMs on Multi-Turn Clarification for Task Formulation in Computational Science](https://mlsanity.com/paper/2605.18630/) (2026)
- [GTA: Generating Long-Horizon Tasks for Web Agents at Scale](https://mlsanity.com/paper/2605.29218/) (2026)
- [Clarify Before Executing: A Self-Evolving Agent for Resolving Intent Asymmetry in 3D Tool Orchestration](https://mlsanity.com/paper/2607.16352/) (2026)
- [Interactive Task Alignment as a POMDP](https://mlsanity.com/paper/2607.16412/) (2026)
- [InsufficiencyBench: Evaluating LLM legal advice on underspecified user queries](https://mlsanity.com/paper/2608.20220/) (2026)
- [Communication Policy Evolution for Proactive LLM Agents](https://mlsanity.com/paper/2606.14314/) (2026)
- [Guideline-as-Oracle: Zero-Annotation Training of an Ophthalmic Telephone Triage Agent](https://mlsanity.com/paper/2608.04772/) (2026)
- [GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting](https://mlsanity.com/paper/2609.12165/) (2026)
- [Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms](https://mlsanity.com/paper/2609.27321/) (2026)
- [Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents](https://mlsanity.com/paper/2609.37236/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly expected to function as collaborative partners, engaging in back-and-forth dialogue to solve complex, ambiguous problems. However, current LLMs often falter in real-world settings, defaulting to passive responses or narrow clarifications when faced with incomplete or under-specified prompts, falling short of proactively gathering the missing information that is crucial for high-quality solutions. In this work, we introduce a new task paradigm: proactive information gathering, where LLMs must identify gaps in the provided context and strategically elicit implicit user knowledge through targeted questions. To systematically study and train this capability, we design a scalable framework that generates partially specified, real-world tasks, masking key information and simulating authentic ambiguity. Within this setup, our core innovation is a reinforcement finetuning strategy that rewards questions that elicit genuinely new, implicit user information -- such as hidden domain expertise or fine-grained requirements -- that would otherwise remain unspoken. Experiments demonstrate that our trained Qwen-2.5-7B model significantly outperforms o3-mini by 18% on automatic evaluation metrics. More importantly, human evaluation reveals that clarification questions and final outlines generated by our model are favored by human annotators by 42% and 28% respectively. Together, these results highlight the value of proactive clarification in elevating LLMs from passive text generators to genuinely collaborative thought partners.