{"total":86,"rows":[{"id":"2410.10781","title":"When Attention Sink Emerges in Language Models: An Empirical View","created":"2024-10-14","cited_by_count":208,"area":"inference"},{"id":"2412.19442","title":"A Survey on Large Language Model Acceleration based on KV Cache Management","created":"2024-12-27","cited_by_count":131,"area":"lr"},{"id":"2407.18003","title":"Keep the Cost Down: A Review on Methods to Optimize LLM' s KV-Cache Consumption","created":"2024-07-25","cited_by_count":97,"area":"inference"},{"id":"2510.00636","title":"Expected Attention: KV Cache Compression by Estimating Attention from Future Queries Distribution","created":"2025-10-01","cited_by_count":74,"area":"inference"},{"id":"2406.12335","title":"Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters","created":"2024-06-18","cited_by_count":73,"area":"inference"},{"id":"2410.01699","title":"Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding","created":"2024-10-02","cited_by_count":66,"area":"inference"},{"id":"2410.15999","title":"Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering","created":"2024-10-21","cited_by_count":63,"area":"mechinterp"},{"id":"2506.15745","title":"InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video Understanding","created":"2025-06-18","cited_by_count":58,"area":"inference"},{"id":"2508.15717","title":"StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding","created":"2025-08-21","cited_by_count":58,"area":"mllm"},{"id":"2504.15364","title":"KeyDiff: Key Similarity-Based KV Cache Eviction for Long-Context LLM Inference in Resource-Constrained Environments","created":"2025-04-21","cited_by_count":51,"area":"inference"},{"id":"2504.02441","title":"Cognitive Memory in Large Language Models","created":"2025-04-03","cited_by_count":29,"area":null},{"id":"2504.17768","title":"The Sparse Frontier: Sparse Attention Trade-offs in Transformer LLMs","created":"2025-04-24","cited_by_count":29,"area":null},{"id":"2503.08727","title":"Training Plug-n-Play Knowledge Modules with Deep Context Distillation","created":"2025-03-11","cited_by_count":24,"area":null},{"id":"2510.07293","title":"AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMs","created":"2025-10-08","cited_by_count":22,"area":"inference"},{"id":"2510.15731","title":"Attention Sinks in Diffusion Language Models","created":"2025-10-17","cited_by_count":20,"area":"diffusion"},{"id":"2412.08890","title":"Lexico: Extreme KV Cache Compression via Sparse Coding over Universal Dictionaries","created":"2024-12-12","cited_by_count":17,"area":"inference"},{"id":"2412.17483","title":"A Silver Bullet or a Compromise for Full Attention? A Comprehensive Study of Gist Token-based Context Compression","created":"2024-12-23","cited_by_count":17,"area":"inference"},{"id":"2510.00231","title":"The Pitfalls of KV Cache Compression","created":"2025-09-30","cited_by_count":15,"area":"inference"},{"id":"2505.20334","title":"Lookahead Q-Cache: Achieving More Consistent KV Cache Eviction via Pseudo Query","created":"2025-05-24","cited_by_count":14,"area":"inference"},{"id":"2510.20707","title":"Mixing Importance with Diversity: Joint Optimization for KV Cache Compression in Large Vision-Language Models","created":"2025-10-23","cited_by_count":14,"area":"mllm"},{"id":"2602.10238","title":"Learning to Evict from Key-Value Cache","created":"2026-02-10","cited_by_count":14,"area":"inference"},{"id":"2502.12082","title":"AdaSplash: Adaptive Sparse Flash Attention","created":"2025-02-17","cited_by_count":12,"area":null},{"id":"2412.12706","title":"More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache Compression","created":"2024-12-17","cited_by_count":10,"area":"inference"},{"id":"2506.17121","title":"Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?","created":"2025-06-20","cited_by_count":10,"area":null},{"id":"2509.24291","title":"Let LLMs Speak Embedding Languages: Generative Text Embeddings via Iterative Contrastive Refinement","created":"2025-09-29","cited_by_count":10,"area":null},{"id":"2505.16175","title":"QuickVideo: Real-Time Long Video Understanding with System Algorithm Co-Design","created":"2025-05-22","cited_by_count":8,"area":null},{"id":"2603.01761","title":"Position: Modular Memory is the Key to Continual Learning Agents","created":"2026-03-02","cited_by_count":8,"area":"lr"},{"id":"2606.03928","title":"Value-Aware Stochastic KV Cache Eviction for Reasoning Models","created":"2026-06-02","cited_by_count":8,"area":"lr"},{"id":"2504.04704","title":"LagKV: Lag-Relative Information of the KV Cache Tells Which Tokens Are Important","created":"2025-04-07","cited_by_count":7,"area":"inference"},{"id":"2506.05410","title":"Homogeneous Keys, Heterogeneous Values: Exploiting Local KV Cache Asymmetry for Long-Context LLMs","created":"2025-06-04","cited_by_count":7,"area":"inference"},{"id":"2509.04377","title":"PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference","created":"2025-09-04","cited_by_count":7,"area":"inference"},{"id":"2412.16187","title":"HashEvict: A Pre-Attention KV Cache Eviction Strategy using Locality-Sensitive Hashing","created":"2024-12-13","cited_by_count":6,"area":"inference"},{"id":"2509.00388","title":"GraphKV: Breaking the Static Selection Paradigm with Graph-Based KV Cache Eviction","created":"2025-08-30","cited_by_count":6,"area":"inference"},{"id":"2512.12008","title":"Hold Onto That Thought: Assessing KV Cache Compression On Reasoning","created":"2025-12-12","cited_by_count":6,"area":"lr"},{"id":"2606.01563","title":"MomentKV: Closing the Directional Gap in KV Cache Eviction for Long-Context Inference","created":"2026-06-01","cited_by_count":6,"area":"inference"},{"id":"2408.01890","title":"Cross-layer Attention Sharing for Pre-trained Large Language Models","created":"2024-08-04","cited_by_count":5,"area":"inference"},{"id":"2504.01281","title":"Scaling Test-Time Inference with Policy-Optimized, Dynamic Retrieval-Augmented Generation via KV Caching and Decoding","created":"2025-04-02","cited_by_count":5,"area":"rag"},{"id":"2507.11273","title":"KV-Latent: Dimensional-level KV Cache Reduction with Frequency-aware Rotary Positional Embedding","created":"2025-07-15","cited_by_count":5,"area":"inference"},{"id":"2604.25975","title":"Rethinking KV Cache Eviction via a Unified Information-Theoretic Objective","created":"2026-04-28","cited_by_count":5,"area":"lr"},{"id":"2410.13166","title":"An Evolved Universal Transformer Memory","created":"2024-10-17","cited_by_count":4,"area":null},{"id":"2503.02812","title":"Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression","created":"2025-03-04","cited_by_count":4,"area":"inference"},{"id":"2503.04973","title":"Beyond RAG: Task-Aware KV Cache Compression for Comprehensive Knowledge Reasoning","created":"2025-03-06","cited_by_count":4,"area":"lr"},{"id":"2504.07109","title":"OSCAR: Online Soft Compression And Reranking","created":"2025-03-17","cited_by_count":4,"area":"rag"},{"id":"2509.10798","title":"Judge Q: Trainable Queries for Optimized Information Retention in KV Cache Eviction","created":"2025-09-13","cited_by_count":4,"area":"inference"},{"id":"2512.17452","title":"KV Admission: Learning What to Write for Efficient Long-Context LLM Inference","created":"2025-12-19","cited_by_count":4,"area":null},{"id":"2512.21859","title":"TimeBill: Time-Budgeted Inference for Large Language Models","created":"2025-12-26","cited_by_count":4,"area":null},{"id":"2603.19664","title":"The Residual Stream Is All You Need: On the Redundancy of the KV Cache in Transformer Inference","created":"2026-03-20","cited_by_count":4,"area":"inference"},{"id":"2607.08057","title":"Towards Efficient Large Language Model Serving: A Survey on System-Aware KV Cache Optimization","created":"2026-07-09","cited_by_count":4,"area":"inference"},{"id":"2503.08461","title":"FastCache: Optimizing Multimodal LLM Serving through Lightweight KV-Cache Compression Framework","created":"2025-03-11","cited_by_count":3,"area":"inference"},{"id":"2505.24179","title":"SALE : Low-bit Estimation for Efficient Sparse Attention in Long-context LLM Prefilling","created":"2025-05-30","cited_by_count":3,"area":null},{"id":"2507.04221","title":"Context Tuning for In-Context Optimization","created":"2025-07-06","cited_by_count":3,"area":null},{"id":"2607.28699","title":"WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization","created":"2026-07-30","cited_by_count":3,"area":"inference"},{"id":"2408.04107","title":"FDC: Fast KV Dimensionality Compression for Efficient LLM Inference","created":"2024-08-07","cited_by_count":2,"area":"inference"},{"id":"2409.11650","title":"Art and Science of Quantizing Large-Scale Models: A Comprehensive Overview","created":"2024-09-18","cited_by_count":2,"area":null},{"id":"2410.12850","title":"RecurFormer: Not All Transformer Heads Need Self-Attention","created":"2024-10-10","cited_by_count":2,"area":null},{"id":"2512.07503","title":"SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image Generation","created":"2025-12-08","cited_by_count":2,"area":"inference"},{"id":"2602.08343","title":"ManifoldKV: Training-Free KV Cache Compression via Euclidean Outlier Detection","created":"2026-02-09","cited_by_count":2,"area":"inference"},{"id":"2603.27819","title":"KVSculpt: KV Cache Compression as Distillation","created":"2026-03-29","cited_by_count":2,"area":"inference"},{"id":"2604.18002","title":"Neural Garbage Collection: Learning to Forget while Learning to Reason","created":"2026-04-20","cited_by_count":2,"area":"lr"},{"id":"2605.26678","title":"NestedKV: Nested Memory Routing for Long-Context KV Cache Compression","created":"2026-05-26","cited_by_count":2,"area":"inference"},{"id":"2606.26472","title":"Epiphany-Aware KV Cache Eviction Without the Attention Matrix","created":"2026-06-25","cited_by_count":2,"area":"lr"},{"id":"2608.01631","title":"Does Accuracy Equal Evidence? Reasoning Faithfulness under KV Cache Compression","created":"2026-08-03","cited_by_count":2,"area":"inference"},{"id":"2608.05326","title":"QEvict: Recoverable Quantized KV Eviction for Attention-Drift-Robust Long-Context Decoding","created":"2026-08-05","cited_by_count":2,"area":null},{"id":"2507.19906","title":"CaliDrop: KV Cache Compression with Calibration","created":"2025-07-26","cited_by_count":1,"area":"inference"},{"id":"2601.18383","title":"Dynamic Thinking-Token Selection for Efficient Reasoning in Large Reasoning Models","created":"2026-01-26","cited_by_count":1,"area":"lr"},{"id":"2603.12038","title":"Slow-Fast Inference: Training-Free Inference Acceleration via Within-Sentence Support Stability","created":"2026-03-12","cited_by_count":1,"area":"inference"},{"id":"2605.07234","title":"Reformulating KV Cache Eviction Problem for Long-Context LLM Inference","created":"2026-05-08","cited_by_count":1,"area":"inference"},{"id":"2606.00620","title":"FlowNar: Scalable Streaming Narration for Long-Form Videos","created":"2026-05-30","cited_by_count":1,"area":"mllm"},{"id":"2609.10266","title":"KVShareArena: KV-Cache Reuse Across Contexts and Model Checkpoints","created":"2026-09-09","cited_by_count":1,"area":"inference"},{"id":"2610.00412","title":"EchoPress: Query-Agnostic KV Cache Pruning via Virtual Context Reconstruction","created":"2026-09-30","cited_by_count":1,"area":"inference"},{"id":"2610.03198","title":"KV$^2$: A Self-Refining KV Cache","created":"2026-10-02","cited_by_count":1,"area":"inference"},{"id":"2603.27960","title":"Towards Efficient Large Vision-Language Models: A Comprehensive Survey on Inference Strategies","created":"2026-03-30","cited_by_count":0,"area":"lr"},{"id":"2605.20600","title":"Head-Aware Key-Value Compression for Efficient Autoregressive Image Generation","created":"2026-05-20","cited_by_count":0,"area":null},{"id":"2605.21649","title":"EntmaxKV: Support-Aware Decoding for Entmax Attention","created":"2026-05-20","cited_by_count":0,"area":"inference"},{"id":"2605.29873","title":"Moment-KV: Momentum-Based Decode-Time KV Cache Compression for Long Generation","created":"2026-05-28","cited_by_count":0,"area":"inference"},{"id":"2606.07604","title":"Contribution Weights: A Geometrical Analysis of Self-Attention Transformers","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2606.10944","title":"Express Language Modeling","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2607.11942","title":"How Query Visibility Changes KV-Cache Compression Rankings: A Matched-Budget Audit","created":"2026-07-11","cited_by_count":0,"area":"inference"},{"id":"2608.28293","title":"A Probabilistic Interpretation of KV Cache Eviction","created":"2026-08-28","cited_by_count":0,"area":"inference"},{"id":"2609.06663","title":"ECOKV: Geometry-Aware KV Cache Eviction via Complementary Diversity Metrics","created":"2026-09-06","cited_by_count":0,"area":"mllm"},{"id":"2609.08131","title":"Jacap: Robust KV Cache Eviction via Jacobian-Based Nonlinear Information Capacity Preservation","created":"2026-09-08","cited_by_count":0,"area":"inference"},{"id":"2609.23314","title":"ValueDiff: Value-Geometric KV Cache Eviction for Sink-Suppressed LLMs","created":"2026-09-20","cited_by_count":0,"area":"inference"},{"id":"2609.27470","title":"DeltaS: Reading the Gated Linear Attention State for KV Cache Eviction in Streaming Video","created":"2026-09-23","cited_by_count":0,"area":"inference"},{"id":"2609.33334","title":"When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache Compression","created":"2026-09-27","cited_by_count":0,"area":"inference"},{"id":"2610.14754","title":"GateSight: Recurrent Memory Anticipates Attention Memory for Query-Agnostic KV Eviction in Hybrid LLMs","created":"2026-10-10","cited_by_count":0,"area":null},{"id":"2610.17191","title":"FreeSolo: Linear-time Sparse Attention with Zero-Shot Length Extrapolation","created":"2026-10-12","cited_by_count":0,"area":null}],"next_cursor":null}