In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917541495963648 |
|---|---|
| author | Choi, Youngbin Lee, Minjong Moon, Saemi Cho, Seunghyuk Chung, Chaehyeon Park, MoonJeong Kim, Dongwoo |
| author_facet | Choi, Youngbin Lee, Minjong Moon, Saemi Cho, Seunghyuk Chung, Chaehyeon Park, MoonJeong Kim, Dongwoo |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_00777 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration Choi, Youngbin Lee, Minjong Moon, Saemi Cho, Seunghyuk Chung, Chaehyeon Park, MoonJeong Kim, Dongwoo Machine Learning 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. |
| title | In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.00777 |