Image de couverture de livre

LoopGuard: Breaking Self-Reinforcing Attention Loops via Dynamic KV Cache Intervention

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Xu, Dongjie, Wu, Hao, Shi, Weijie, Cui, Yue, Liu, Yuanjun, Li, Jiawei, Ma, Haolun, Liu, An, Zhu, Jia, Xu, Jiajie
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Cognitive-Uncertainty Guided Knowledge Distillation for Accurate Classification of Student Misconceptions
par: Liu, Qirui, et autres
Publié: (2026)

Escher-Loop: Mutual Evolution by Closed-Loop Self-Referential Optimization
par: Liu, Ziyang, et autres
Publié: (2026)

PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
par: Cai, Zefan, et autres
Publié: (2024)

Adaptation Method for Misinformation Identification
par: Chen, Yangping, et autres
Publié: (2025)

Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management
par: Zheng, Haoyu, et autres
Publié: (2026)

Sparse Attention across Multiple-context KV Cache
par: Cao, Ziyi, et autres
Publié: (2025)

LoopRPT: Reinforcement Pre-Training for Looped Language Models
par: Tang, Guo, et autres
Publié: (2026)

AttentionPredictor: Temporal Patterns Matter for KV Cache Compression
par: Yang, Qingyue, et autres
Publié: (2025)

SQuat: Subspace-orthogonal KV Cache Quantization
par: Wang, Hao, et autres
Publié: (2025)

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads
par: He, Xingyang, et autres
Publié: (2025)

E3-Rewrite: Learning to Rewrite SQL for Executability, Equivalence,and Efficiency
par: Xu, Dongjie, et autres
Publié: (2025)

In-context KV-Cache Eviction for LLMs via Attention-Gate
par: Zeng, Zihao, et autres
Publié: (2024)

SSR3D-LLM: Structured Spatial Reasoning via Latent Steps for Fine-Grained Grounding in Unified 3D-LLMs
par: Li, Jiawei, et autres
Publié: (2026)

Cross-Self KV Cache Pruning for Efficient Vision-Language Inference
par: Pei, Xiaohuan, et autres
Publié: (2024)

Homogeneous Keys, Heterogeneous Values: Exploiting Local KV Cache Asymmetry for Long-Context LLMs
par: Cui, Wanyun, et autres
Publié: (2025)

GraphKV: Breaking the Static Selection Paradigm with Graph-Based KV Cache Eviction
par: Li, Xuelin, et autres
Publié: (2025)

Zero-Shot Cellular Trajectory Map Matching
par: Shi, Weijie, et autres
Publié: (2025)

G-KV: Decoding-Time KV Cache Eviction with Global Attention
par: Liao, Mengqi, et autres
Publié: (2025)

AhaKV: Adaptive Holistic Attention-Driven KV Cache Eviction for Efficient Inference of Large Language Models
par: Gu, Yifeng, et autres
Publié: (2025)

Lookahead Q-Cache: Achieving More Consistent KV Cache Eviction via Pseudo Query
par: Wang, Yixuan, et autres
Publié: (2025)

Adaptive KV-Cache Compression without Manually Setting Budget
par: Tang, Chenxia, et autres
Publié: (2025)

ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partition
par: Ye, Lu, et autres
Publié: (2024)

Breaking the Loop: Detecting and Mitigating Denial-of-Service Vulnerabilities in Large Language Models
par: Yu, Junzhe, et autres
Publié: (2025)

PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference
par: Yang, Dongjie, et autres
Publié: (2024)

Ship Detection in SAR Images with Human-in-the-Loop
par: Jia, Hecheng, et autres
Publié: (2024)

Compressing KV Cache for Long-Context LLM Inference with Inter-Layer Attention Similarity
par: Ma, Da, et autres
Publié: (2024)

RefreshKV: Updating Small KV Cache During Long-form Generation
par: Xu, Fangyuan, et autres
Publié: (2024)

ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
par: Liu, Xiang, et autres
Publié: (2025)

KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache Sharing
par: Yang, Yifei, et autres
Publié: (2024)

Some Bounds on the Energy of Graphs with Self-Loops regarding $λ_{1}$ and $λ_{n}$
par: Li, Minghua, et autres
Publié: (2024)

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

DioR: Adaptive Cognitive Detection and Contextual Retrieval Optimization for Dynamic Retrieval-Augmented Generation
par: Guo, Hanghui, et autres
Publié: (2025)

Towards Threshold-Free KV Cache Pruning
par: Ni, Xuanfan, et autres
Publié: (2025)

Transactional Attention: Semantic Sponsorship for KV-Cache Retention
par: Basu, Abhinaba
Publié: (2026)

Beyond KV Caching: Shared Attention for Efficient LLMs
par: Liao, Bingli, et autres
Publié: (2024)

Circular Reasoning: Understanding Self-Reinforcing Loops in Large Reasoning Models
par: Duan, Zenghao, et autres
Publié: (2026)

XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization
par: Tomar, Aditya, et autres
Publié: (2025)

Beyond Homogeneous Attention: Memory-Efficient LLMs via Fourier-Approximated KV Cache
par: Liu, Xiaoran, et autres
Publié: (2025)

ACR: Adaptive Context Refactoring via Context Refactoring Operators for Multi-Turn Dialogue
par: Shen, Jiawei, et autres
Publié: (2026)

Eigen Attention: Attention in Low-Rank Space for KV Cache Compression
par: Saxena, Utkarsh, et autres
Publié: (2024)