{"total":258,"rows":[{"id":"2505.24298","title":"AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning","created":"2025-05-30","cited_by_count":250,"area":"lr"},{"id":"2502.01776","title":"Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity","created":"2025-02-03","cited_by_count":217,"area":"diffusion"},{"id":"2505.18875","title":"Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation","created":"2025-05-24","cited_by_count":106,"area":"diffusion"},{"id":"2506.06266","title":"Cartridges: Lightweight and general-purpose long context representations via self-study","created":"2025-06-06","cited_by_count":86,"area":"inference"},{"id":"2507.11948","title":"Kevin: Multi-Turn RL for Generating CUDA Kernels","created":"2025-07-16","cited_by_count":78,"area":null},{"id":"2506.19852","title":"Radial Attention: $O(n\\log n)$ Sparse Attention with Energy Decay for Long Video Generation","created":"2025-06-24","cited_by_count":69,"area":"diffusion"},{"id":"2502.02770","title":"Twilight: Adaptive Attention Sparsity with Hierarchical Top-$p$ Pruning","created":"2025-02-04","cited_by_count":49,"area":null},{"id":"2511.02230","title":"Continuum: Efficient and Robust Multi-Turn LLM Agent Scheduling with KV Cache Time-to-Live","created":"2025-11-04","cited_by_count":44,"area":"agents"},{"id":"2502.14051","title":"RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression","created":"2025-02-19","cited_by_count":41,"area":"inference"},{"id":"2507.17773","title":"MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation","created":"2025-07-20","cited_by_count":38,"area":null},{"id":"2604.14141","title":"LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstruction","created":"2026-04-15","cited_by_count":37,"area":null},{"id":"2608.30320","title":"On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability","created":"2026-08-31","cited_by_count":35,"area":"moe"},{"id":"2509.23202","title":"Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantization","created":"2025-09-27","cited_by_count":32,"area":null},{"id":"2601.00227","title":"FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems","created":"2026-01-01","cited_by_count":25,"area":null},{"id":"2502.12216","title":"Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs","created":"2025-02-17","cited_by_count":24,"area":"inference"},{"id":"2602.19128","title":"K-Search: LLM Kernel Generation via Co-Evolving Intrinsic World Model","created":"2026-02-22","cited_by_count":24,"area":"wm"},{"id":"2511.15915","title":"AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization","created":"2025-11-19","cited_by_count":23,"area":"agents"},{"id":"2602.05885","title":"Dr. Kernel: Reinforcement Learning Done Right for Triton Kernel Generations","created":"2026-02-05","cited_by_count":22,"area":"lr"},{"id":"2505.07203","title":"PrefillOnly: An Inference Engine for Prefill-only Workloads in Large Language Model Applications","created":"2025-05-12","cited_by_count":21,"area":"inference"},{"id":"2504.07347","title":"Throughput-Optimal Scheduling Algorithms for LLM Inference and AI Agents","created":"2025-04-10","cited_by_count":18,"area":"agents"},{"id":"2602.01801","title":"Fast Autoregressive Video Diffusion and World Models with Temporal Cache Compression and Sparse Attention","created":"2026-02-02","cited_by_count":18,"area":"wm"},{"id":"2501.18107","title":"Scaling Inference-Efficient Language Models","created":"2025-01-30","cited_by_count":16,"area":null},{"id":"2502.02534","title":"Adaptive Self-improvement LLM Agentic System for ML Library Development","created":"2025-02-04","cited_by_count":16,"area":"agents"},{"id":"2603.12938","title":"Thinking in Streaming Video","created":"2026-03-13","cited_by_count":16,"area":null},{"id":"2603.18567","title":"SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding","created":"2026-03-19","cited_by_count":16,"area":"inference"},{"id":"2512.22219","title":"MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs","created":"2025-12-22","cited_by_count":15,"area":null},{"id":"2601.11580","title":"Speculative Decoding: Performance or Illusion?","created":"2025-12-31","cited_by_count":15,"area":"inference"},{"id":"2603.03251","title":"Speculative Speculative Decoding","created":"2026-03-03","cited_by_count":15,"area":"inference"},{"id":"2502.01068","title":"FastKV: Decoupling of Context Reduction and KV Cache Compression for Prefill-Decoding Acceleration","created":"2025-02-03","cited_by_count":14,"area":"inference"},{"id":"2507.08944","title":"Optimizing Sequential Multi-Step Tasks with Parallel LLM Agents","created":"2025-07-11","cited_by_count":14,"area":"agents"},{"id":"2505.02922","title":"RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference","created":"2025-05-05","cited_by_count":13,"area":"inference"},{"id":"2507.06608","title":"Nexus:Proactive Intra-GPU Disaggregation of Prefill and Decode in LLM Serving","created":"2025-07-09","cited_by_count":13,"area":"inference"},{"id":"2505.18658","title":"Robustness in Large Language Models: A Survey of Mitigation Strategies and Evaluation Metrics","created":"2025-05-24","cited_by_count":12,"area":null},{"id":"2506.05333","title":"Kinetics: Rethinking Test-Time Scaling Laws","created":"2025-06-05","cited_by_count":12,"area":"lr"},{"id":"2602.04789","title":"Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attention","created":"2026-02-04","cited_by_count":12,"area":"diffusion"},{"id":"2601.15727","title":"Towards Automated Kernel Generation in the Era of LLMs","created":"2026-01-22","cited_by_count":11,"area":"agents"},{"id":"2603.16104","title":"Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective","created":"2026-03-17","cited_by_count":11,"area":"agents"},{"id":"2501.14808","title":"HyGen: Efficient LLM Serving via Elastic Online-Offline Request Co-location","created":"2025-01-15","cited_by_count":10,"area":"inference"},{"id":"2512.01278","title":"Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding","created":"2025-12-01","cited_by_count":10,"area":"lr"},{"id":"2511.11571","title":"Optimizing Mixture of Block Attention","created":"2025-11-14","cited_by_count":9,"area":null},{"id":"2512.02895","title":"MindGPT-4ov: An Enhanced MLLM via a Multi-Stage Post-Training Paradigm","created":"2025-12-02","cited_by_count":9,"area":"mllm"},{"id":"2503.04596","title":"LLM Applications: Current Paradigms and the Next Frontier","created":"2025-03-06","cited_by_count":8,"area":"agents"},{"id":"2506.03275","title":"Chipmunk: Training-Free Acceleration of Diffusion Transformers with Dynamic Column-Sparse Deltas","created":"2025-06-03","cited_by_count":8,"area":"diffusion"},{"id":"2508.18255","title":"Hermes 4 Technical Report","created":"2025-08-25","cited_by_count":8,"area":"lr"},{"id":"2511.17560","title":"$A^3$: Attention-Aware Accurate KV Cache Fusion for Fast Large Language Model Serving","created":"2025-11-13","cited_by_count":8,"area":"rag"},{"id":"2604.03143","title":"TokenDance: Scaling Multi-Agent LLM Serving via Collective KV Cache Sharing","created":"2026-04-03","cited_by_count":8,"area":"inference"},{"id":"2502.04077","title":"AttentionPredictor: Temporal Patterns Matter for KV Cache Compression","created":"2025-02-06","cited_by_count":7,"area":"inference"},{"id":"2505.22913","title":"Mustafar: Promoting Unstructured Sparsity for KV Cache Pruning in LLM Inference","created":"2025-05-28","cited_by_count":7,"area":"inference"},{"id":"2506.02572","title":"HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference","created":"2025-06-03","cited_by_count":7,"area":"inference"},{"id":"2601.04426","title":"XGrammar-2: Dynamic and Efficient Structured Generation Engine for Agentic LLMs","created":"2026-01-07","cited_by_count":7,"area":"agents"},{"id":"2602.07223","title":"Vegas: Self-Speculative Decoding with Verification-Guided Sparse Attention","created":"2026-02-06","cited_by_count":7,"area":"inference"},{"id":"2503.20533","title":"Accelerate Parallelizable Reasoning via Parallel Decoding within One Sequence","created":"2025-03-26","cited_by_count":6,"area":"lr"},{"id":"2508.01002","title":"Optimal Scheduling Algorithms for LLM Inference: Theory and Practice","created":"2025-08-01","cited_by_count":6,"area":"inference"},{"id":"2510.08544","title":"SPAD: Specialized Prefill and Decode Hardware for Disaggregated LLM Inference","created":"2025-10-09","cited_by_count":6,"area":null},{"id":"2510.09883","title":"DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoning","created":"2025-10-10","cited_by_count":6,"area":"lr"},{"id":"2510.14719","title":"Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References","created":"2025-10-16","cited_by_count":6,"area":null},{"id":"2604.18348","title":"AdaCluster: Adaptive Query-Key Clustering for Sparse Attention in Video Generation","created":"2026-04-20","cited_by_count":6,"area":"diffusion"},{"id":"2604.25899","title":"Pythia: Exploiting Workflow Predictability for Efficient Agent-Native LLM Serving","created":"2026-04-28","cited_by_count":6,"area":"agents"},{"id":"2506.01969","title":"FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs","created":"2025-05-13","cited_by_count":5,"area":"lr"},{"id":"2508.17219","title":"TokenLake: A Unified Segment-level Prefix Cache Pool for Fine-grained Elastic Long-Context LLM Serving","created":"2025-08-24","cited_by_count":5,"area":"inference"},{"id":"2511.22880","title":"Serving Heterogeneous LoRA Adapters in Distributed LLM Inference Systems","created":"2025-11-28","cited_by_count":5,"area":"inference"},{"id":"2604.10539","title":"IceCache: Memory-efficient KV-cache Management for Long-Sequence LLMs","created":"2026-04-12","cited_by_count":5,"area":"inference"},{"id":"2604.19241","title":"UniEP: Unified Expert-Parallel MoE MegaKernel for LLM Training","created":"2026-04-21","cited_by_count":5,"area":"moe"},{"id":"2605.18071","title":"KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference","created":"2026-05-18","cited_by_count":5,"area":"inference"},{"id":"2605.19269","title":"CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs","created":"2026-05-19","cited_by_count":5,"area":null},{"id":"2606.10651","title":"Kwai Keye-VL-2.0 Technical Report","created":"2026-06-09","cited_by_count":5,"area":"agents"},{"id":"2607.20468","title":"InferenceBench: A Benchmark for Open-Ended LLM Inference Optimization by AI Agents","created":"2026-05-20","cited_by_count":5,"area":"agents"},{"id":"2608.12629","title":"CAKE: Compiler-Agent Co-Design for Frontier Kernel Evolution","created":"2026-08-12","cited_by_count":5,"area":null},{"id":"2504.04104","title":"SpecPipe: Accelerating Pipeline Parallelism-based LLM Inference with Speculative Decoding","created":"2025-04-05","cited_by_count":4,"area":"inference"},{"id":"2508.09001","title":"Retrospective Sparse Attention for Efficient Long-Context Generation","created":"2025-08-12","cited_by_count":4,"area":"inference"},{"id":"2509.16518","title":"FG-Attn: Leveraging Fine-Grained Sparse Attention in Video Diffusion Models","created":"2025-09-20","cited_by_count":4,"area":"diffusion"},{"id":"2510.08726","title":"Neptune: Advanced ML Operator Fusion for Locality and Parallelism on GPUs","created":"2025-10-09","cited_by_count":4,"area":null},{"id":"2510.13602","title":"NOSA: Native and Offloadable Sparse Attention","created":"2025-10-15","cited_by_count":4,"area":"inference"},{"id":"2510.18245","title":"Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs","created":"2025-10-21","cited_by_count":4,"area":null},{"id":"2510.18413","title":"Adamas: Hadamard Sparse Attention for Efficient Long-Context Inference","created":"2025-10-21","cited_by_count":4,"area":null},{"id":"2602.02108","title":"Out of the Memory Barrier: A Highly Memory Efficient Training System for LLMs with Million-Token Contexts","created":"2026-02-02","cited_by_count":4,"area":"inference"},{"id":"2602.21224","title":"Make Every Draft Count: Hidden State based Speculative Decoding","created":"2026-02-02","cited_by_count":4,"area":"inference"},{"id":"2603.26380","title":"Switch Attention: Towards Dynamic and Fine-grained Hybrid Transformers","created":"2026-03-27","cited_by_count":4,"area":null},{"id":"2604.13327","title":"Event Tensor: A Unified Abstraction for Compiling Dynamic Megakernel","created":"2026-04-14","cited_by_count":4,"area":null},{"id":"2605.06068","title":"VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?","created":"2026-05-07","cited_by_count":4,"area":"agents"},{"id":"2605.13111","title":"Pyramid Forcing: Head-Aware Pyramid KV Cache Policy for High-Quality Long Video Generation","created":"2026-05-13","cited_by_count":4,"area":"inference"},{"id":"2605.20868","title":"Runtime-Certified Bounded-Error Quantized Attention","created":"2026-05-20","cited_by_count":4,"area":"inference"},{"id":"2605.21312","title":"Frontier: Towards Comprehensive and Accurate LLM Inference Simulation","created":"2026-05-20","cited_by_count":4,"area":"inference"},{"id":"2605.30571","title":"Memory-Bound but Not Bandwidth-Limited: The Physical AI Inference Gap in Batch-1 LLM Decode","created":"2026-05-28","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":"2506.09316","title":"On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention","created":"2025-06-11","cited_by_count":3,"area":null},{"id":"2509.24328","title":"Speculative Verification: Exploiting Information Gain to Refine Speculative Decoding","created":"2025-09-29","cited_by_count":3,"area":"inference"},{"id":"2510.16276","title":"What Limits Agentic Systems Efficiency?","created":"2025-10-18","cited_by_count":3,"area":"lr"},{"id":"2511.06838","title":"P3-LLM: An Integrated NPU-PIM Accelerator for Edge LLM Inference Using Hybrid Numerical Formats","created":"2025-11-10","cited_by_count":3,"area":null},{"id":"2511.11581","title":"The Anatomy of a Triton Attention Kernel","created":"2025-10-07","cited_by_count":3,"area":null},{"id":"2511.13274","title":"KForge: Program Synthesis for Diverse AI Hardware Accelerators","created":"2025-11-17","cited_by_count":3,"area":null},{"id":"2601.13684","title":"HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference","created":"2026-01-20","cited_by_count":3,"area":"inference"},{"id":"2601.19092","title":"Axe: A Simple Unified Layout Abstraction for Machine Learning Compilers","created":"2026-01-27","cited_by_count":3,"area":null},{"id":"2602.00269","title":"VoxServe: Streaming-Centric Serving System for Speech Language Models","created":"2026-01-30","cited_by_count":3,"area":"inference"},{"id":"2603.05931","title":"A Persistent-State Dataflow Accelerator for Memory-Bound Linear Attention Decode on FPGA","created":"2026-03-06","cited_by_count":3,"area":"inference"},{"id":"2603.07236","title":"HY-WU (Part I): An Extensible Functional Neural Memory Framework and An Instantiation in Text-Guided Image Editing","created":"2026-03-07","cited_by_count":3,"area":null},{"id":"2603.15854","title":"FlashSampling: Fast and Memory-Efficient Exact Sampling","created":"2026-03-16","cited_by_count":3,"area":null},{"id":"2605.17170","title":"TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks","created":"2026-05-16","cited_by_count":3,"area":"agents"},{"id":"2605.30748","title":"Chatterbox-Flash: Prior-Calibrated Block Diffusion for Streaming Zero-Shot TTS","created":"2026-05-29","cited_by_count":3,"area":null},{"id":"2606.10537","title":"Prefilling-dLLM: Predictive Prefilling for Long-Context Inference in Diffusion Language Models","created":"2026-06-09","cited_by_count":3,"area":"diffusion"}],"next_cursor":"100"}