{"total":55,"rows":[{"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":"2510.13999","title":"REAP the Experts: Why Pruning Prevails for One-Shot MoE compression","created":"2025-10-15","cited_by_count":44,"area":"moe"},{"id":"2411.02335","title":"Sparsing Law: Towards Large Language Models with Greater Activation Sparsity","created":"2024-11-04","cited_by_count":22,"area":null},{"id":"2504.20966","title":"Softpick: No Attention Sink, No Massive Activations with Rectified Softmax","created":"2025-04-29","cited_by_count":22,"area":null},{"id":"2504.19449","title":"R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference","created":"2025-04-28","cited_by_count":21,"area":"inference"},{"id":"2504.18415","title":"BitNet v2: Native 4-bit Activations with Hadamard Transformation for 1-bit LLMs","created":"2025-04-25","cited_by_count":18,"area":null},{"id":"2411.04965","title":"BitNet a4.8: 4-bit Activations for 1-bit LLMs","created":"2024-11-07","cited_by_count":16,"area":null},{"id":"2503.16672","title":"Accelerating Transformer Inference and Training with 2:4 Activation Sparsity","created":"2025-03-20","cited_by_count":9,"area":null},{"id":"2505.05950","title":"FloE: On-the-Fly MoE Inference on Memory-constrained GPU","created":"2025-05-09","cited_by_count":9,"area":"moe"},{"id":"2506.16500","title":"SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity","created":"2025-06-19","cited_by_count":9,"area":null},{"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":"2502.00840","title":"Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense","created":"2025-02-02","cited_by_count":7,"area":null},{"id":"2509.00454","title":"Universal Properties of Activation Sparsity in Modern Large Language Models","created":"2025-08-30","cited_by_count":7,"area":null},{"id":"2502.01378","title":"CE-LoRA: Computation-Efficient LoRA Fine-Tuning for Language Models","created":"2025-02-03","cited_by_count":6,"area":null},{"id":"2505.14884","title":"Polar Sparsity: High Throughput Batched LLM Inferencing with Scalable Contextual Sparsity","created":"2025-05-20","cited_by_count":6,"area":null},{"id":"2506.20194","title":"DuoGPT: Training-free Dual Sparsity through Activation-aware Pruning in LLMs","created":"2025-06-25","cited_by_count":6,"area":null},{"id":"2507.01299","title":"La RoSA: Enhancing LLM Efficiency via Layerwise Rotated Sparse Activation","created":"2025-07-02","cited_by_count":6,"area":null},{"id":"2506.06644","title":"Spark Transformer: Reactivating Sparsity in FFN and Attention","created":"2025-06-07","cited_by_count":5,"area":null},{"id":"2507.19608","title":"DeltaLLM: Delta-Key Sparse Attention for Edge LLMs","created":"2025-07-25","cited_by_count":5,"area":null},{"id":"2504.08378","title":"Scaling Up On-Device LLMs via Active-Weight Swapping Between DRAM and Flash","created":"2025-04-11","cited_by_count":4,"area":null},{"id":"2510.26843","title":"CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs","created":"2025-10-30","cited_by_count":4,"area":"inference"},{"id":"2505.19427","title":"WINA: Weight Informed Neuron Activation for Accelerating Large Language Model Inference","created":"2025-05-26","cited_by_count":3,"area":"moe"},{"id":"2412.05644","title":"Mixture of Hidden-Dimensions Transformer","created":"2024-12-07","cited_by_count":2,"area":null},{"id":"2607.08601","title":"It Takes a MAESTRO To Prune Bad Experts","created":"2026-07-09","cited_by_count":2,"area":"moe"},{"id":"2508.14302","title":"GLASS: Global-Local Aggregation for Inference-time Sparsification of LLMs","created":"2025-08-19","cited_by_count":1,"area":null},{"id":"2511.04477","title":"Enabling Dynamic Sparsity in Quantized LLM Inference","created":"2025-11-06","cited_by_count":1,"area":null},{"id":"2511.18692","title":"VLM in a flash: I/O-Efficient Sparsification of Vision-Language Model via Neuron Chunking","created":"2025-11-24","cited_by_count":1,"area":"mllm"},{"id":"2601.11676","title":"HALO: Semantic-Aware Distributed LLM Inference in Lossy Edge Network","created":"2026-01-16","cited_by_count":1,"area":null},{"id":"2601.14792","title":"Robustness of Mixtures of Experts to Feature Noise","created":"2026-01-21","cited_by_count":1,"area":"moe"},{"id":"2602.14452","title":"WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation Sparsity","created":"2026-02-16","cited_by_count":1,"area":"inference"},{"id":"2603.12272","title":"ActTail: Global Activation Sparsity in Large Language Models","created":"2026-02-18","cited_by_count":1,"area":null},{"id":"2605.17659","title":"Bug or Feature$^2$: Weight Drift, Activation Sparsity and Spikes","created":"2026-05-17","cited_by_count":1,"area":null},{"id":"2606.23419","title":"GRINQH: Graded Input-based Quantization Hierarchy for Efficient LLM Generation","created":"2026-06-22","cited_by_count":1,"area":null},{"id":"2503.23817","title":"MVDRAM: Enabling GeMV Execution in Unmodified DRAM for Low-Bit LLM Acceleration","created":"2025-03-31","cited_by_count":0,"area":null},{"id":"2506.23225","title":"Masked Gated Linear Unit","created":"2025-06-29","cited_by_count":0,"area":null},{"id":"2509.22358","title":"Stochastic activations","created":"2025-09-26","cited_by_count":0,"area":null},{"id":"2512.12744","title":"Resting Neurons, Active Insights: Robustifying Activation Sparsity in LLMs via Spontaneity","created":"2025-12-14","cited_by_count":0,"area":null},{"id":"2602.06822","title":"POP: Online Structural Pruning Enables Efficient Inference of Large Foundation Models","created":"2026-02-06","cited_by_count":0,"area":null},{"id":"2603.14110","title":"SVD Contextual Sparsity Predictors for Fast LLM Inference","created":"2026-03-14","cited_by_count":0,"area":null},{"id":"2603.23198","title":"Sparser, Faster, Lighter Transformer Language Models","created":"2026-03-24","cited_by_count":0,"area":null},{"id":"2605.03109","title":"Gated Subspace Inference for Transformer Acceleration","created":"2026-05-04","cited_by_count":0,"area":null},{"id":"2605.10875","title":"Compute Where it Counts: Self Optimizing Language Models","created":"2026-05-11","cited_by_count":0,"area":"inference"},{"id":"2605.25612","title":"Towards the Connection between Activation Sparsity and Flat Minima","created":"2026-05-25","cited_by_count":0,"area":null},{"id":"2605.26632","title":"RT-Lynx: Putting GEMM Sparsity in the Right Place for Diffusion Models","created":"2026-05-26","cited_by_count":0,"area":"diffusion"},{"id":"2606.00365","title":"SPARQLe: Sub-Precision Activation Representation for Quantized LLM Inference","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2607.25504","title":"At-the-Roofline Sparse Tensor Contractions on Vector Processors for Transformer Inference","created":"2026-07-28","cited_by_count":0,"area":null},{"id":"2608.02995","title":"SparSEEty: Extracting Tokens from Sparsity-Exploiting LLM Serving Systems via Deterministic Side Channels","created":"2026-08-04","cited_by_count":0,"area":"inference"},{"id":"2608.12953","title":"Unifying Depth and Width Pruning for LLMs via Binary Knapsack Optimization","created":"2026-08-13","cited_by_count":0,"area":null},{"id":"2608.14003","title":"Batch-wise Adaptive Pruning: Periodic Neuron Activation-Aware Weight Pruning for Language Reasoning Model","created":"2026-08-14","cited_by_count":0,"area":"lr"},{"id":"2608.30439","title":"Event-Driven Language Models with Sparse Neural Activity for Neuromorphic Hardware","created":"2026-08-31","cited_by_count":0,"area":"inference"},{"id":"2609.03079","title":"LeanStream: A Speculate-and-Refine Streaming Framework for Efficient on-Device LLM Inference","created":"2026-09-02","cited_by_count":0,"area":null},{"id":"2609.31975","title":"Model Casting and Low-Parameter Gating: Towards More Sparsely Activated FFNs","created":"2026-09-25","cited_by_count":0,"area":null},{"id":"2609.33889","title":"Where Activation Sparsity and KV-Cache Sparsity Cross in LLM Decoding","created":"2026-09-27","cited_by_count":0,"area":"inference"},{"id":"2610.01763","title":"TopK-Guided: Adaptive, Budget-Aware Activation Sparsity for Efficient LLM Inference","created":"2026-10-01","cited_by_count":0,"area":"inference"},{"id":"2610.02598","title":"Activation Sparsity with Weight Approximation for Faster LLM Decoding on Offloaded Weights","created":"2026-10-01","cited_by_count":0,"area":null}],"next_cursor":null}