让用户直接修改模型输出，提升纠错效率与协作体验。
In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration
- 用户直接编辑模型前序回复，模型从修改处继续生成。
- 在五项推理任务中表现优于传统多轮反馈，且用更少的token完成。
- 专家实测显示满意度更高、疲劳感更低，适合专业用户协作。
LLM生成的草稿常包含细微的事实或逻辑错误，但以往研究显示模型难以可靠地整合旨在修正这些错误的多轮反馈。本文提出‘在位反馈’（in-place feedback）机制：用户直接编辑模型先前的回复，模型从修改后的上下文继续生成。该方法在五个推理密集型基准测试中均优于标准多轮反馈，且所需标记数更少。细粒度分析表明，其能更可靠地应用修正并传递至后续推理。专家用户研究进一步验证：参与者对最终输出满意度更高，疲劳感显著降低；混合使用在位与多轮反馈在所有评估维度上得分最高。结果表明，直接编辑错误是更有效的专家-大模型协作范式。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。