Does Execution Require Target KV Fidelity? A Mixed-Fidelity KV Runtime for LLM Serving
cs.LG
Submitted: 2026-09-27
Updated: 2026-09-27
Terminology
Sources
- PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
- The Llama 3 Herd of Models
- Lynx: Progressive Speculative Quantization for accelerating KV Transfer in Long-Context Inference
- Benchmarking Neural Network Robustness to Common Corruptions and Perturbations
- GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
- Zipage: Maintain High Request Concurrency for LLM Reasoning through Compressed PagedAttention
- LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Fast Model Editing at Scale
- KV-Compress: Paged KV-Cache Compression with Variable Compression Rates per Attention Head
- FastSwitch: Optimizing Context Switching Efficiency in Fairness-aware Large Language Model Serving
- Coordinated Scheduling for MoE LLM Serving
- Qwen3 Technical Report
- Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors
- Semantic Robustness Certification for Vision-Language 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