English

In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration

Machine Learning 2026-05-29 v2

Abstract

LLM-generated drafts often contain subtle factual or logical errors, yet prior work shows that models struggle to reliably integrate multi-turn feedback aimed at fixing them. We propose in-place feedback, an interaction paradigm in which the user directly edits the model's previous response and the model continues generation from the edited context. In-place feedback consistently outperforms standard multi-turn feedback across five reasoning-intensive benchmarks while requiring fewer tokens, and our fine-grained analysis shows that it applies corrections more reliably and propagates them to subsequent reasoning. A user study with domain experts refining LLM-generated summaries corroborates these findings: participants report higher final-output satisfaction and substantially lower fatigue with in-place feedback, and a mixed strategy combining in-place and multi-turn feedback scores highest on every measured dimension. These results suggest that editing errors directly is a more effective paradigm for expert-LLM collaboration.

Keywords

Cite

@article{arxiv.2510.00777,
  title  = {In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration},
  author = {Youngbin Choi and Minjong Lee and Saemi Moon and Seunghyuk Cho and Chaehyeon Chung and MoonJeong Park and Dongwoo Kim},
  journal= {arXiv preprint arXiv:2510.00777},
  year   = {2026}
}

Comments

42pages

R2 v1 2026-07-01T06:10:21.149Z