CommitKV: Lifecycle-Aware KV Cache Compression via Commit Transitions for Multi-Turn Agents
Weizhong Huang, Jinchao Zhang, Xiawu Zheng
cs.LG
Submitted: 2026-08-08
Updated: 2026-08-11
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Phi-4-reasoning Technical Report
- gpt-oss-120b & gpt-oss-20b Model Card
- InternLM2 Technical Report
- Value-Aware Stochastic KV Cache Eviction for Reasoning Models
- xbench: Tracking Agents Productivity Scaling with Profession-Aligned Real-World Evaluations
- Exploring Large Language Model based Intelligent Agents: Definitions, Methods, and Prospects
- Large Language Model based Multi-Agents: A Survey of Progress and Challenges
- Self-GC: Self-Governing Context for Long-Horizon LLM Agents
- Determining Layer-wise Sparsity for Large Language Models Through a Theoretical Perspective
- GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
- SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
- EpiCache: Episodic KV Cache Management for Long-Term Conversation on Resource-Constrained Environments
- IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference
- HiMemVLN: Enhancing Reliability of Open-Source Zero-Shot Vision-and-Language Navigation with Hierarchical Memory System
- PhotoCraft: Agentic Reasoning with Hierarchical Self-Evolving Memory for Deep Image Search
- Flow caching for autoregressive video generation
- TriAttention: Efficient Long Reasoning with Trigonometric KV Compression
- The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Survey
- MemDecay: Region-Aware KV Cache Eviction for Efficient LLM Agent Inference
- Keep the Cost Down: A Review on Methods to Optimize LLM' s KV-Cache Consumption
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks