{"total":36,"rows":[{"id":"2402.02057","title":"Break the Sequential Dependency of LLM Inference Using Lookahead Decoding","created":"2024-02-03","cited_by_count":298,"area":"inference"},{"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":"2407.15891","title":"RazorAttention: Efficient KV Cache Compression Through Retrieval Heads","created":"2024-07-22","cited_by_count":100,"area":"inference"},{"id":"2412.01827","title":"RandAR: Decoder-only Autoregressive Visual Generation in Random Orders","created":"2024-12-02","cited_by_count":83,"area":"inference"},{"id":"2312.04916","title":"EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism","created":"2023-12-08","cited_by_count":69,"area":"inference"},{"id":"2312.05516","title":"Stateful Large Language Model Serving with Pensieve","created":"2023-12-09","cited_by_count":61,"area":"inference"},{"id":"2404.18911","title":"Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting","created":"2024-04-29","cited_by_count":58,"area":"inference"},{"id":"2410.20771","title":"MrT5: Dynamic Token Merging for Efficient Byte-level Language Models","created":"2024-10-28","cited_by_count":28,"area":"inference"},{"id":"2502.07578","title":"PIM Is All You Need: A CXL-Enabled GPU-Free System for Large Language Model Inference","created":"2025-02-11","cited_by_count":22,"area":"inference"},{"id":"2402.00518","title":"EE-Tuning: An Economical yet Scalable Solution for Tuning Early-Exit Large Language Models","created":"2024-02-01","cited_by_count":13,"area":"inference"},{"id":"2502.13652","title":"C2T: A Classifier-Based Tree Construction Method in Speculative Decoding","created":"2025-02-19","cited_by_count":9,"area":"inference"},{"id":"2505.22135","title":"RAD: Redundancy-Aware Distillation for Hybrid Models via Self-Speculative Decoding","created":"2025-05-28","cited_by_count":8,"area":"inference"},{"id":"2506.10443","title":"MNN-LLM: A Generic Inference Engine for Fast Large Language Model Deployment on Mobile Devices","created":"2025-06-12","cited_by_count":8,"area":"inference"},{"id":"2410.15252","title":"Lossless KV Cache Compression to 2%","created":"2024-10-20","cited_by_count":7,"area":"inference"},{"id":"2502.10659","title":"Pushing up to the Limit of Memory Bandwidth and Capacity Utilization for Efficient LLM Decoding on Embedded FPGA","created":"2025-02-15","cited_by_count":7,"area":"inference"},{"id":"2508.10395","title":"XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization","created":"2025-08-14","cited_by_count":4,"area":"inference"},{"id":"2606.07878","title":"Still: Amortized KV Cache Compaction in a Single Forward Pass","created":"2026-06-05","cited_by_count":4,"area":"inference"},{"id":"2606.09659","title":"End-to-End Context Compression at Scale","created":"2026-06-08","cited_by_count":4,"area":"inference"},{"id":"2512.14661","title":"Focus: A Streaming Concentration Architecture for Efficient Vision-Language Models","created":"2025-12-16","cited_by_count":3,"area":"inference"},{"id":"2406.11016","title":"Optimized Speculative Sampling for GPU Hardware Accelerators","created":"2024-06-16","cited_by_count":2,"area":"inference"},{"id":"2507.09768","title":"Knowing When to Quit: Probabilistic Early Exits for Speech Separation","created":"2025-07-13","cited_by_count":2,"area":"inference"},{"id":"2601.19611","title":"Explicit Multi-head Attention for Inter-head Interaction in Large Language Models","created":"2026-01-27","cited_by_count":2,"area":"inference"},{"id":"2602.18196","title":"RAT+: Train Dense, Infer Sparse -- Recurrence Augmented Attention for Dilated Inference","created":"2026-02-20","cited_by_count":2,"area":"inference"},{"id":"2603.02188","title":"Multi-Head Low-Rank Attention","created":"2026-03-02","cited_by_count":2,"area":"inference"},{"id":"2606.20543","title":"SSD: Spatially Speculative Decoding Accelerates Autoregressive Image Generation","created":"2026-06-18","cited_by_count":2,"area":"inference"},{"id":"2502.17728","title":"LLM Inference Acceleration via Efficient Operation Fusion","created":"2025-02-24","cited_by_count":1,"area":"inference"},{"id":"2508.00370","title":"EdgeInfinite-Instruct: Bridging SFT-Based Optimization and NPU-Level Efficiency for Edge Devices","created":"2025-08-01","cited_by_count":1,"area":"inference"},{"id":"2608.06849","title":"Autonomy-of-Heads: Data-Free Sparse Attention from Frozen Query-Key Geometry","created":"2026-08-07","cited_by_count":1,"area":"inference"},{"id":"2511.21408","title":"Subjective Depth and Timescale Transformers: Learning Where and When to Compute","created":"2025-11-26","cited_by_count":0,"area":"inference"},{"id":"2602.02163","title":"Reg4Pru: Regularisation Through Random Token Routing for Token Pruning","created":"2026-02-02","cited_by_count":0,"area":"inference"},{"id":"2603.04797","title":"Hardware-Software Co-design for 3D-DRAM-based LLM Serving Accelerator","created":"2026-03-05","cited_by_count":0,"area":"inference"},{"id":"2608.22110","title":"How Weight Encoding Affects Language Model Placement and Performance on the Apple Neural Engine","created":"2026-08-22","cited_by_count":0,"area":"inference"},{"id":"2608.23296","title":"Sigmoid Attention as a Better Substrate for Learned KV Cache Eviction","created":"2026-08-24","cited_by_count":0,"area":"inference"},{"id":"2609.13285","title":"Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction","created":"2026-09-08","cited_by_count":0,"area":"inference"},{"id":"2610.14475","title":"EdgeLLM-Nano: Auditing Polar-Guided Key-Cache Compression and Consistency-Anchored Pruning on a 33.6 MB On-Device Language Model","created":"2026-10-10","cited_by_count":0,"area":"inference"},{"id":"2610.17202","title":"MLAfication: Efficiently Enabling High-Fidelity MLA for Modern LLMs","created":"2026-10-12","cited_by_count":0,"area":"inference"}],"next_cursor":null}