OVAL: Output-Aware Local Page Bases for KV Cache Retrieval

arXiv:2610.06686 · cs.LG · Submitted 2026-10-05 · Read on arXiv

cs.LG

Submitted: 2026-10-05

Updated: 2026-10-05

Code: https://github.com/Ashkan13776/oval-kv

Terminology

• KV cache — 1× in this paper · explained in 3 episodes

Sources

• Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis
• LOCKS: Page-Local Compact Key Summaries for Efficient Long-Context Decoding
• UNIQUE: Universal Top-k Sparse Attention for Training-free Inference and Sparsity-aware Training
• DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
• FreeKV: Boosting KV Cache Retrieval for Efficient LLM Inference
• TokenButler: Token Importance is Predictable
• SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs
• Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
• LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level Mathematical Reasoning
• Qwen2.5 Technical Report
• Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement
• The Llama 3 Herd of 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: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks