Image de couverture de livre

Intelligence Degradation in Long-Context LLMs: Critical Threshold Determination via Natural Length Distribution Analysis

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Wang, Weiwei, Min, Jiyong, Zou, Weijie
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Solving LLM Repetition Problem in Production: A Comprehensive Study of Multiple Solutions
par: Wang, Weiwei, et autres
Publié: (2025)

Abusive text transformation using LLMs
par: Chandra, Rohitash, et autres
Publié: (2025)

DoubleDipper: Improving Long-Context LLMs via Context Recycling
par: Cattan, Arie, et autres
Publié: (2024)

On Many-Shot In-Context Learning for Long-Context Evaluation
par: Zou, Kaijian, et autres
Publié: (2024)

Intrinsic Entropy of Context Length Scaling in LLMs
par: Shi, Jingzhe, et autres
Publié: (2025)

TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection
par: Wu, Wei, et autres
Publié: (2024)

Why Does the Effective Context Length of LLMs Fall Short?
par: An, Chenxin, et autres
Publié: (2024)

ParallelComp: Parallel Long-Context Compressor for Length Extrapolation
par: Xiong, Jing, et autres
Publié: (2025)

Squeezed Attention: Accelerating Long Context Length LLM Inference
par: Hooper, Coleman, et autres
Publié: (2024)

Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data
par: Deng, Haoran, et autres
Publié: (2025)

Context Discipline and Performance Correlation: Analyzing LLM Performance and Quality Degradation Under Varying Context Lengths
par: Ponnusamy, Ahilan Ayyachamy Nadar, et autres
Publié: (2025)

ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
par: Wu, Xixi, et autres
Publié: (2025)

Infinite Retrieval: Attention Enhanced LLMs in Long-Context Processing
par: Ye, Xiaoju, et autres
Publié: (2025)

Long-Short Alignment for Effective Long-Context Modeling in LLMs
par: Du, Tianqi, et autres
Publié: (2025)

LongEmotion: Measuring Emotional Intelligence of Large Language Models in Long-Context Interaction
par: Liu, Weichu, et autres
Publié: (2025)

Efficient Length-Generalizable Attention via Causal Retrieval for Long-Context Language Modeling
par: Hu, Xiang, et autres
Publié: (2024)

Understanding the RoPE Extensions of Long-Context LLMs: An Attention Perspective
par: Zhong, Meizhi, et autres
Publié: (2024)

DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs
par: Zhou, Xiabin, et autres
Publié: (2024)

Mixture of In-Context Experts Enhance LLMs' Long Context Awareness
par: Lin, Hongzhan, et autres
Publié: (2024)

LIFT: A Novel Framework for Enhancing Long-Context Understanding of LLMs via Long Input Fine-Tuning
par: Mao, Yansheng, et autres
Publié: (2025)

FlashPrefill: Instantaneous Pattern Discovery and Thresholding for Ultra-Fast Long-Context Prefilling
par: Fan, Qihang, et autres
Publié: (2026)

UIO-LLMs: Unbiased Incremental Optimization for Long-Context LLMs
par: Li, Wenhao, et autres
Publié: (2024)

LongReD: Mitigating Short-Text Degradation of Long-Context Large Language Models via Restoration Distillation
par: Dong, Zican, et autres
Publié: (2025)

Scaling Long-Horizon LLM Agent via Context-Folding
par: Sun, Weiwei, et autres
Publié: (2025)

OWL: Overcoming Window Length-Dependence in Speculative Decoding for Long-Context Inputs
par: Lee, Jaeseong, et autres
Publié: (2025)

LongGenBench: Benchmarking Long-Form Generation in Long Context LLMs
par: Wu, Yuhao, et autres
Publié: (2024)

Understanding Textual Capability Degradation in Speech LLMs via Parameter Importance Analysis
par: Wang, Chao, et autres
Publié: (2025)

Understanding LLM Performance Degradation in Multi-Instance Processing: The Roles of Instance Count and Context Length
par: Chen, Jingxuan, et autres
Publié: (2026)

Towards a Holistic and Automated Evaluation Framework for Multi-Level Comprehension of LLMs in Book-Length Contexts
par: Deng, Jiaqi, et autres
Publié: (2025)

LongLLaDA: Unlocking Long Context Capabilities in Diffusion LLMs
par: Liu, Xiaoran, et autres
Publié: (2025)

LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression
par: Jiang, Huiqiang, et autres
Publié: (2023)

Length-MAX Tokenizer for Language Models
par: Dong, Dong, et autres
Publié: (2025)

MathHay: An Automated Benchmark for Long-Context Mathematical Reasoning in LLMs
par: Wang, Lei, et autres
Publié: (2024)

CTkvr: KV Cache Retrieval for Long-Context LLMs via Centroid then Token Indexing
par: Lu, Kuan, et autres
Publié: (2025)

Are Long-LLMs A Necessity For Long-Context Tasks?
par: Qian, Hongjin, et autres
Publié: (2024)

On Stable Long-Form Generation: Benchmarking and Mitigating Length Volatility
par: He, Zhitao, et autres
Publié: (2026)

Sentence-Anchored Gist Compression for Long-Context LLMs
par: Tarasov, Dmitrii, et autres
Publié: (2025)

Joint Enhancement of Relational Reasoning for Long-Context LLMs
par: Chen, Zhirui, et autres
Publié: (2025)

Long Context vs. RAG for LLMs: An Evaluation and Revisits
par: Li, Xinze, et autres
Publié: (2024)

LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K
par: Yuan, Tao, et autres
Publié: (2024)