大模型
arxiv.org ·
面向大语言模型的演化上下文参数化研究
论文形式化提出 Memory Updating with Sequential Evolution（MUSE）任务并构建 MUSE-bench，用于评估上下文持续更新时的更新吸收与未受影响信息保持能力。作者提出免训练方法 PLUME：构建全局更新表示，激活记忆证据形成局部参数视图，并在解码时自适应融合两者预测。在 MUSE-bench 上报告平均 ROUGE-L Recall 相对提升 29.9%、LLM-as-a-Judge 相对提升 54.9%，代码已公开。
AI推荐理由 arXiv 新论文，提出 MUSE 任务与 MUSE-bench 基准，并给出免训练方法 PLUME，用于在上下文持续更新时保留全局状态并调整记忆证据贡献；摘要报告在 MUSE-bench 上 ROUGE-L Recall 相对提升 29.9%、LLM-as-a-Judge 相对提升 54.9%，并称代码已公开。信息密度较高、来源可查，但结果均为作者自报，尚未见独立复现，且基准与任务为新提出，实际泛化能力待验证。
72
正文
Computer Science > Computation and Language
arXiv:2609.14168 (cs)
[Submitted on 12 Sep 2026]
Title:Towards Evolving Context Parameterization for Large Language Models
Authors:Xiaobing Shi, Zherui Li, Yiming Jiang, Kun Wang, Yufei Guo
View a PDF of the paper titled Towards Evolving Context Parameterization for Large Language Models, by Xiaobing Shi and 4 other authors
View PDF
HTML (experimental)
Abstract:Context parameterization enables large language models (LLMs) to internalize contexts into reusable model parameters, avoiding repeated processing across subsequent queries. However, existing methods typically assume static contexts and lack explicit mechanisms for distinguishing validity states under continual updates. To study this real-world scenario, we formalized the Memory Updating with Sequential Evolution (MUSE) task and constructed MUSE-bench to evaluate update incorporation and unaffected-information preservation. The resulting challenge requires preserving the global state while adjusting the contribution of memory evidence. Motivated by this, we proposed PLUME, a training-free method that constructs a global update representation, activates memory evidence to form a local parameter view, and adaptively integrates their predictions during decoding. Comprehensive evaluation on MUSE-bench demonstrated PLUME's effectiveness in sequential evolution settings, yielding relative improvements of 29.9% in average ROUGE-L Recall and 54.9% in LLM-as-a-Judge. Our codes are available at: this https URL.
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as:
arXiv:2609.14168 [cs.CL]
(or
arXiv:2609.14168v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.14168
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Zherui Li [view email]
[v1]
Sat, 12 Sep 2026 21:55:33 UTC (4,256 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Towards Evolving Context Parameterization for Large Language Models, by Xiaobing Shi and 4 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-09
Change to browse by:
cs
cs.AI
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?)