Meta-Reflection: A Feedback-Free Reflection Learning Framework
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912160685228032 |
|---|---|
| author | Wang, Yaoke Zhu, Yun Bao, Xintong Zhang, Wenqiao Dai, Suyang Chen, Kehan Li, Wenqiang Huang, Gang Tang, Siliang Zhuang, Yueting |
| author_facet | Wang, Yaoke Zhu, Yun Bao, Xintong Zhang, Wenqiao Dai, Suyang Chen, Kehan Li, Wenqiang Huang, Gang Tang, Siliang Zhuang, Yueting |
| contents | Despite the remarkable capabilities of large language models (LLMs) in natural language understanding and reasoning, they often display undesirable behaviors, such as generating hallucinations and unfaithful reasoning. A prevalent strategy to mitigate these issues is the use of reflection, which refines responses through an iterative process. However, while promising, reflection heavily relies on high-quality external feedback and requires iterative multi-agent inference processes, thus hindering its practical application. In this paper, we propose Meta-Reflection, a novel feedback-free reflection mechanism that necessitates only a single inference pass without external feedback. Motivated by the human ability to remember and retrieve reflections from past experiences when encountering similar problems, Meta-Reflection integrates reflective insights into a codebook, allowing the historical insights to be stored, retrieved, and used to guide LLMs in problem-solving. To thoroughly investigate and evaluate the practicality of Meta-Reflection in real-world scenarios, we introduce an industrial e-commerce benchmark named E-commerce Customer Intent Detection (ECID). Extensive experiments conducted on both public datasets and the ECID benchmark highlight the effectiveness and efficiency of our proposed approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_13781 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Meta-Reflection: A Feedback-Free Reflection Learning Framework Wang, Yaoke Zhu, Yun Bao, Xintong Zhang, Wenqiao Dai, Suyang Chen, Kehan Li, Wenqiang Huang, Gang Tang, Siliang Zhuang, Yueting Computation and Language Artificial Intelligence Despite the remarkable capabilities of large language models (LLMs) in natural language understanding and reasoning, they often display undesirable behaviors, such as generating hallucinations and unfaithful reasoning. A prevalent strategy to mitigate these issues is the use of reflection, which refines responses through an iterative process. However, while promising, reflection heavily relies on high-quality external feedback and requires iterative multi-agent inference processes, thus hindering its practical application. In this paper, we propose Meta-Reflection, a novel feedback-free reflection mechanism that necessitates only a single inference pass without external feedback. Motivated by the human ability to remember and retrieve reflections from past experiences when encountering similar problems, Meta-Reflection integrates reflective insights into a codebook, allowing the historical insights to be stored, retrieved, and used to guide LLMs in problem-solving. To thoroughly investigate and evaluate the practicality of Meta-Reflection in real-world scenarios, we introduce an industrial e-commerce benchmark named E-commerce Customer Intent Detection (ECID). Extensive experiments conducted on both public datasets and the ECID benchmark highlight the effectiveness and efficiency of our proposed approach. |
| title | Meta-Reflection: A Feedback-Free Reflection Learning Framework |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2412.13781 |