按年被提到
20262026 · 2 次2026
大模型与语言模型›大模型推理与思维链·arXiv:2602.16066 · cs.AI

Improving Interactive In-Context Learning from Natural Language Feedback

Martin Klissarov, Jonathan Cook, Diego Antognini, Hao Sun, Jingling Li, Natasha Jaques, Claudiu Musat, Edward Grefenstette · 2026
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-02-20 · 最近 2026-04-08
2不同的人
2活跃月份
2讨论串
2被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Adapting one's thought process based on corrective feedback is an essential ability in human learning, particularly in collaborative settings. In contrast, the current large language model training paradigm relies heavily on modeling vast, static corpora. While effective for knowledge acquisition, it overlooks the interactive feedback loops essential for models to adapt dynamically to their context. In this work, we propose a framework that treats this interactive in-context learning ability not as an emergent property, but as a distinct, trainable skill. We introduce a scalable method that transforms single-turn verifiable tasks into multi-turn didactic interactions driven by information asymmetry. We first show that current flagship models struggle to integrate corrective feedback on hard reasoning tasks. We then demonstrate that models trained with our approach dramatically improve the ability to interactively learn from language feedback. More specifically, the multi-turn performance of a smaller model nearly reaches that of a model an order of magnitude larger. We also observe robust out-of-distribution generalization: interactive training on math problems transfers to diverse domains like coding, puzzles and maze navigation. Our qualitative analysis suggests that this improvement is due to an enhanced in-context plasticity. Finally, we show that this paradigm offers a unified path to self-improvement. By training the model to predict the teacher's critiques, effectively modeling the feedback environment, we convert this external signal into an internal capability, allowing the model to self-correct even without a teacher.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论revv00 · 2026-04

“Interesting paper, it reminds me how we human improving our math problem solving capability. But those efficient guys use few problems(and reasoning trajectories) to build their solving skills, how to build this effective human training process into LLM's?”

Improving Interactive In-Context Learning from Natural Language Feedback原评论 ↗
被提交为 story · 2 次