# Precise Information Control in Long-Form Text Generation

arXiv: https://arxiv.org/abs/2506.06589 · HTML: https://mlsanity.com/paper/2506.06589/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Jun 2025.

## Built on by (most cited first, 6 of 6)

- [Characterizing Deep Research: A Benchmark and Formal Definition](https://mlsanity.com/paper/2508.04183/) (2025)
- [HalluHard: A Hard Multi-Turn Hallucination Benchmark](https://mlsanity.com/paper/2602.01031/) (2026)
- [Deep-Reporter: Deep Research for Grounded Multimodal Long-Form Generation](https://mlsanity.com/paper/2604.10741/) (2026)
- [Do Reasoning LLMs Refuse What They Infer in Long Contexts?](https://mlsanity.com/paper/2602.08874/) (2026)
- [Who is the Agent to Blame? Localizing Faithfulness and Citation Mistakes in Agentic Deep Research](https://mlsanity.com/paper/2608.24306/) (2026)
- [IS-CoT: Breaking the Long-form Generation Collapse via Interleaved Structural Thinking](https://mlsanity.com/paper/2606.09709/) (2026)

## Abstract (arXiv)

A central challenge in language models (LMs) is faithfulness hallucination: the generation of information unsubstantiated by input context. To study this problem, we propose Precise Information Control (PIC), a new task formulation that requires models to generate long-form outputs grounded in a provided set of short self-contained statements, without adding any unsupported ones. PIC includes a full setting that tests a model's ability to include exactly all input claims, and a partial setting that requires the model to selectively incorporate only relevant claims. We present PIC-Bench, a benchmark of eight long-form generation tasks (e.g., summarization, biography generation) adapted to the PIC setting, where LMs are supplied with well-formed, verifiable input claims. Our evaluation of a range of open and proprietary LMs on PIC-Bench reveals that, surprisingly, state-of-the-art LMs still hallucinate against user-provided input in over 70% of generations. To alleviate this lack of faithfulness, we introduce a post-training framework that uses a weakly supervised preference data construction method to train an 8B PIC-LM with stronger PIC ability--improving from 69.1% to 91.0% F1 in the full PIC setting. When integrated into end-to-end factual generation pipelines, PIC-LM improves exact match recall by 17.1% on ambiguous QA with retrieval, and factual precision by 30.5% on a birthplace fact-checking task, underscoring the potential of precisely grounded generation.