paper-with-me

홈 › Papers

ReflectEvo: Improving Meta Introspection of Small LLMs by Learning Self-Reflection

2025-05-22 · Jiaqi Li, Xinyi Dong, Yang Liu, Zhizhuo Yang, Quansen Wang, Xiaobo Wang, Songchun Zhu, Zixia Jia, Zilong Zheng

We present a novel pipeline, ReflectEvo, to demonstrate that small language models (SLMs) can enhance meta introspection through reflection learning. This process iteratively generates self-reflection for self-training, fostering a continuous and self-evolving process. Leveraging this pipeline, we construct ReflectEvo-460k, a large-scale, comprehensive, self-generated reflection dataset with broadened instructions and diverse multi-domain tasks. Building upon this dataset, we demonstrate the effectiveness of reflection learning to improve SLMs' reasoning abilities using SFT and DPO with remarkable performance, substantially boosting Llama-3 from 52.4% to 71.2% and Mistral from 44.4% to 71.1%. It validates that ReflectEvo can rival or even surpass the reasoning capability of the three prominent open-sourced models on BIG-bench without distillation from superior models or fine-grained human annotation. We further conduct a deeper analysis of the high quality of self-generated reflections and their impact on error localization and correction. Our work highlights the potential of continuously enhancing the reasoning performance of SLMs through iterative reflection learning in the long run.

📄 PDF Abstract BibTeX arXiv:2505.16475

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…
DPO 설명 없음

Similar Papers 제목 키워드 기반

Self-Reference in Large Language Models: The Introspection Threshold for Recursive Self-Improvement

2026-07-05 · Jiang Zhang, Bing Yuan, Qian Zhang arxiv

The pursuit of self-evolving AI raises a critical question: when is autonomous self-improvement sustainable rather than degenerative? Drawing an analogy to von Neumann's complexity threshold for self-reproducing automata…

Me, Myself, and $π$ : Evaluating and Explaining LLM Introspection

2026-03-17 · Atharv Naphade, Samarth Bhargav, Sean Lim, Mcnair Shah arxiv

A hallmark of human intelligence is Introspection-the ability to assess and reason about one's own cognitive processes. Introspection has emerged as a promising but contested capability in large language models (LLMs). H…

Language Models Fail to Introspect About Their Knowledge of Language

2025-03-10 · Siyuan Song, Jennifer Hu, Kyle Mahowald

There has been recent interest in whether large language models (LLMs) can introspect about their own internal states. Such abilities would make LLMs more interpretable, and also validate the use of standard introspectiv…

Sentence

Does It Make Sense to Speak of Introspection in Large Language Models?

2025-06-05 · Iulia M. Comsa, Murray Shanahan

Large language models (LLMs) exhibit compelling linguistic behaviour, and sometimes offer self-reports, that is to say statements about their own nature, inner workings, or behaviour. In humans, such reports are often at…

valid

InnerPond: Fostering Inter-Self Dialogue with a Multi-Agent Approach for Introspection

2026-03-29 · Hayeon Jeon, Dakyeom Ahn, Sunyu Pang, Yunseo Choi 외 arxiv

Introspection is central to identity construction and future planning, yet most digital tools approach the self as a unified entity. In contrast, Dialogical Self Theory (DST) views the self as composed of multiple intern…