大模型
arxiv.org ·
每时每刻更智能：面向持续 LLM 改进的环境驱动动态策略
该 arXiv 论文提出 Dynamic Retrieval-based Policy Generation（DRPG）框架，将基于记忆的检索与动态策略生成器结合，利用历史数据和环境反馈生成任务特定策略，以应对 LLM 持续适应演化任务与环境的挑战。作者在 text-to-SQL、问答、医疗诊断和 Python 编程等六个基准上，使用七种专有与开源 LLM 评估，称 DRPG 在多数数据集和模型上优于强基线；分析显示其策略生成对检索策略鲁棒、无需先前策略连续性，并可用更小或跨家族模型作为低成本策略生成器，同时指出策略级指导的收益取决于任务特征。
AI推荐理由 arXiv 新提交论文（2026-09-15），提出 DRPG 框架，将基于记忆的检索与动态策略生成器结合，利用历史数据和环境反馈为持续 LLM 改进生成任务特定策略。在 text-to-SQL、问答、医疗诊断、Python 编程等六个基准、七种专有与开源 LLM 上评估，摘要称在多数数据集和模型上优于强基线，并分析了策略生成对检索策略的鲁棒性、无需先前策略连续性、可用更小或跨家族模型作为低成本策略生成器，以及策略级指导收益依赖任务特征。信息密度较高，属可验证的研究工作；但仅有摘要，无代码或复现证据，效果声明未经独立验证，故未给更高分。
72
正文
Computer Science > Computation and Language
arXiv:2609.16800 (cs)
[Submitted on 15 Sep 2026]
Title:Smarter by the Moment: Environment-Driven Dynamic Policies for Continual LLM Improvement
Authors:Ting-Wei Chang, Po-Chun Chen, Hen-Hsen Huang, Hsin-Hsi Chen
View a PDF of the paper titled Smarter by the Moment: Environment-Driven Dynamic Policies for Continual LLM Improvement, by Ting-Wei Chang and 3 other authors
View PDF
HTML (experimental)
Abstract:Large Language Models (LLMs) have achieved remarkable progress across diverse domains, but continual adaptation to evolving tasks and environments remains a key challenge. Existing memory-augmented approaches retrieve individual past examples as direct references, but do not explicitly synthesize actionable strategies from them, causing the same types of errors to recur. We propose Dynamic Retrieval-based Policy Generation (DRPG), a framework that integrates memory-based retrieval with a dynamic policy generator, leveraging historical data and environment feedback to produce task-specific policies for continual LLM improvement. We evaluate DRPG across six benchmarks spanning text-to-SQL, question answering, medical diagnosis, and Python programming, using seven LLMs from both proprietary and open-weight families. DRPG outperforms strong baselines across most datasets and models. Further analysis demonstrates that DRPG's policy generation is robust to retrieval strategy, operates effectively without prior policy continuity, and can leverage smaller or cross-family models as cost-efficient policy generators. We also find that the benefit of policy-level guidance depends on task characteristics, offering practical insights into when and under what conditions this mechanism is most effective.
Comments:
Subjects:
Computation and Language (cs.CL)
Cite as:
arXiv:2609.16800 [cs.CL]
(or
arXiv:2609.16800v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.16800
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Ting-Wei Chang [view email]
[v1]
Tue, 15 Sep 2026 08:06:17 UTC (309 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Smarter by the Moment: Environment-Driven Dynamic Policies for Continual LLM Improvement, by Ting-Wei Chang and 3 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-09
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)