EpiKV: Epiphany-Aware KV Cache Eviction Without the Attention Matrix

arXiv:2606.26472 · cs.LG, cs.CL · Submitted 2026-06-25 · Read on arXiv

cs.LG, cs.CL

Submitted: 2026-06-25

Updated: 2026-09-29

Terminology

• Trace — 28× in this paper · explained in 2 episodes
• KV cache — 22× in this paper · explained in 3 episodes
• Accuracy — 17× in this paper · explained in 1 episode
• Decode — 13× in this paper · explained in 2 episodes
• Reasoning Traces — 8× in this paper · explained in 3 episodes
• Benchmark — 6× in this paper · explained in 2 episodes
• Quantization — 5× in this paper · explained in 12 episodes
• Sparsity — 4× in this paper · explained in 4 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Attention Entropy — 2× in this paper · explained in 2 episodes
• Calibration — 2× in this paper · explained in 7 episodes
• In International Conference on Learning Representations — 2× in this paper
• Agentic Reasoning — 1× in this paper · explained in 1 episode
• Embeddings — 1× in this paper · explained in 2 episodes
• Filler Tokens — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Speculative Decoding — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Training Verifiers to Solve Math Word Problems
• CAOTE: KV Cache Selection for LLMs via Attention Output Error-Based Token Eviction
• AhaKV: Adaptive Holistic Attention-Driven KV Cache Eviction for Efficient Inference of Large Language Models
• SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
• LagKV: Lag-Relative Information of the KV Cache Tells Which Tokens Are Important
• LongFlow: Efficient KV Cache Compression for Reasoning Models

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: Local-Consistency Optimisation for Bayesian Neural Networks
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks