ICRL: Learning to Internalize Self-Critique with Reinforcement Learning
The paper presents a novel framework called ICRL, which aims to enhance the self-improvement capabilities of language model-based agents. By jointly training a solver and a critic, ICRL enables the model to internalize critique and improve its performance without relying on external feedback. The results demonstrate significant performance gains on various reasoning tasks, indicating the effectiveness of this approach.
- ▪ICRL stands for Learning to Internalize Self-Critique with Reinforcement Learning.
- ▪The framework trains a solver and a critic from a shared backbone to enhance the model's ability to self-improve.
- ▪Results show average performance gains of 6.4 points on agentic tasks and 7.0 points on mathematical reasoning tasks.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.15224 |
| Publication time | Mon, 18 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-18T04:04:54.418Z |
| Last seen | 2026-05-18T04:04:54.418Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | 8uNqKJmf5Mlt |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.15224 (cs) [Submitted on 13 May 2026] Title:ICRL: Learning to Internalize Self-Critique with Reinforcement Learning Authors:Jianbo Lin, Xiaomin Yu, Yi Xin, Yifu Guo, Zhuosong Jiang, Zhongqi Yue, Weishi Wang, Heqing Zou, Chengwei Qin, Hui Xiong View a PDF of the paper titled ICRL: Learning to Internalize Self-Critique with Reinforcement Learning, by Jianbo Lin and 8 other authors View PDF HTML (experimental) Abstract:Large language model-based agents make mistakes, yet critique can often guide the same model toward correct behavior. However, when critique is removed, the model may fail again on the same query, indicating that it has not internalized the critique's guidance into its underlying capability.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
