{"total":151,"rows":[{"id":"2309.17453","title":"Efficient Streaming Language Models with Attention Sinks","created":"2023-09-29","cited_by_count":2186,"area":null},{"id":"2405.16406","title":"SpinQuant: LLM quantization with learned rotations","created":"2024-05-26","cited_by_count":471,"area":"inference"},{"id":"2402.17762","title":"Massive Activations in Large Language Models","created":"2024-02-27","cited_by_count":294,"area":null},{"id":"2505.06708","title":"Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free","created":"2025-05-10","cited_by_count":283,"area":"moe"},{"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":"2503.03321","title":"See What You Are Told: Visual Attention Sink in Large Multimodal Models","created":"2025-03-05","cited_by_count":132,"area":"mllm"},{"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":"2411.13676","title":"Hymba: A Hybrid-head Architecture for Small Language Models","created":"2024-11-20","cited_by_count":99,"area":null},{"id":"2405.00208","title":"A Primer on the Inner Workings of Transformer-based Language Models","created":"2024-04-30","cited_by_count":85,"area":null},{"id":"2402.18158","title":"Evaluating Quantized Large Language Models","created":"2024-02-28","cited_by_count":77,"area":null},{"id":"2403.12544","title":"AffineQuant: Affine Transformation Quantization for Large Language Models","created":"2024-03-19","cited_by_count":67,"area":null},{"id":"2402.09748","title":"Model Compression and Efficient Inference for Large Language Models: A Survey","created":"2024-02-15","cited_by_count":64,"area":null},{"id":"2410.13835","title":"Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs","created":"2024-10-17","cited_by_count":56,"area":null},{"id":"2403.01241","title":"IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact","created":"2024-03-02","cited_by_count":52,"area":null},{"id":"2510.06477","title":"Attention Sinks and Compression Valleys in LLMs are Two Sides of the Same Coin","created":"2025-10-07","cited_by_count":52,"area":null},{"id":"2309.05516","title":"Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs","created":"2023-09-11","cited_by_count":50,"area":null},{"id":"2406.06385","title":"Low-Rank Quantization-Aware Training for LLMs","created":"2024-06-10","cited_by_count":43,"area":null},{"id":"2407.03211","title":"How Does Quantization Affect Multilingual LLMs?","created":"2024-07-03","cited_by_count":43,"area":null},{"id":"2402.02446","title":"LQER: Low-Rank Quantization Error Reconstruction for LLMs","created":"2024-02-04","cited_by_count":41,"area":null},{"id":"2506.08010","title":"Vision Transformers Don't Need Trained Registers","created":"2025-06-09","cited_by_count":41,"area":null},{"id":"2404.03827","title":"Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models","created":"2024-04-04","cited_by_count":38,"area":null},{"id":"2404.03828","title":"Outlier-Efficient Hopfield Layers for Large Transformer-Based Models","created":"2024-04-04","cited_by_count":38,"area":null},{"id":"2410.04962","title":"Activation Scaling for Steering and Interpreting Language Models","created":"2024-10-07","cited_by_count":35,"area":null},{"id":"2512.02010","title":"Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling","created":"2025-12-01","cited_by_count":35,"area":null},{"id":"2405.06001","title":"LLMC: Benchmarking Large Language Model Quantization with a Versatile Compression Toolkit","created":"2024-05-09","cited_by_count":33,"area":"lr"},{"id":"2406.12016","title":"Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization","created":"2024-06-17","cited_by_count":33,"area":null},{"id":"2406.03136","title":"Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models","created":"2024-06-05","cited_by_count":31,"area":null},{"id":"2407.17465","title":"u-$\\mu$P: The Unit-Scaled Maximal Update Parametrization","created":"2024-07-24","cited_by_count":29,"area":null},{"id":"2407.21658","title":"Beat this! Accurate beat tracking without DBN postprocessing","created":"2024-07-31","cited_by_count":27,"area":null},{"id":"2601.14004","title":"Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models","created":"2026-01-20","cited_by_count":27,"area":"mechinterp"},{"id":"2603.05498","title":"The Spike, the Sparse and the Sink: Anatomy of Massive Activations and Attention Sinks","created":"2026-03-05","cited_by_count":27,"area":null},{"id":"2403.06408","title":"What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation","created":"2024-03-11","cited_by_count":23,"area":null},{"id":"2601.22966","title":"A Unified View of Attention and Residual Sinks: Outlier-Driven Rescaling is Essential for Transformer Training","created":"2026-01-30","cited_by_count":23,"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":"2405.14428","title":"Mitigating Quantization Errors Due to Activation Spikes in GLU-Based LLMs","created":"2024-05-23","cited_by_count":21,"area":null},{"id":"2309.17224","title":"Training and inference of large language models using 8-bit floating point","created":"2023-09-29","cited_by_count":20,"area":null},{"id":"2402.04779","title":"StableMask: Refining Causal Masking in Decoder-only Transformer","created":"2024-02-07","cited_by_count":20,"area":null},{"id":"2405.17822","title":"Conv-CoA: Improving Open-domain Question Answering in Large Language Models via Conversational Chain-of-Action","created":"2024-05-28","cited_by_count":19,"area":null},{"id":"2307.00331","title":"Quantization Variation: A New Perspective on Training Transformers with Low-Bit Precision","created":"2023-07-01","cited_by_count":17,"area":null},{"id":"2408.13933","title":"MobileQuant: Mobile-friendly Quantization for On-device Language Models","created":"2024-08-25","cited_by_count":17,"area":null},{"id":"2507.23279","title":"Unveiling Super Experts in Mixture-of-Experts Large Language Models","created":"2025-07-31","cited_by_count":17,"area":"moe"},{"id":"2502.00919","title":"Attention Sinks: A 'Catch, Tag, Release' Mechanism for Embeddings","created":"2025-02-02","cited_by_count":16,"area":null},{"id":"2503.08908","title":"Interpreting the Repeated Token Phenomenon in Large Language Models","created":"2025-03-11","cited_by_count":16,"area":null},{"id":"2506.04985","title":"FPTQuant: Function-Preserving Transforms for LLM Quantization","created":"2025-06-05","cited_by_count":16,"area":null},{"id":"2309.01729","title":"Softmax Bias Correction for Quantized Generative Models","created":"2023-09-04","cited_by_count":15,"area":"diffusion"},{"id":"2407.12397","title":"Mamba-PTQ: Outlier Channels in Recurrent Large Language Models","created":"2024-07-17","cited_by_count":15,"area":"ssm"},{"id":"2407.08044","title":"RoLoRA: Fine-tuning Rotated Outlier-free LLMs for Effective Weight-Activation Quantization","created":"2024-07-10","cited_by_count":13,"area":null},{"id":"2410.05258","title":"Differential Transformer","created":"2024-10-07","cited_by_count":13,"area":null},{"id":"2410.16682","title":"Methods of improving LLM training stability","created":"2024-10-22","cited_by_count":13,"area":null},{"id":"2410.17174","title":"From Attention to Activation: Unravelling the Enigmas of Large Language Models","created":"2024-10-22","cited_by_count":13,"area":null},{"id":"2410.17980","title":"Scaling Stick-Breaking Attention: An Efficient Implementation and In-depth Study","created":"2024-10-23","cited_by_count":13,"area":null},{"id":"2506.19697","title":"Outlier-Safe Pre-Training for Robust 4-Bit Quantization of Large Language Models","created":"2025-06-24","cited_by_count":13,"area":null},{"id":"2410.05265","title":"PrefixQuant: Eliminating Outliers by Prefixed Tokens for Large Language Models Quantization","created":"2024-10-07","cited_by_count":12,"area":null},{"id":"2410.18234","title":"Multi-Draft Speculative Sampling: Canonical Decomposition and Theoretical Limits","created":"2024-10-23","cited_by_count":11,"area":"inference"},{"id":"2603.11487","title":"Attention Sinks Are Provably Necessary in Softmax Transformers: Evidence from Trigger-Conditional Tasks","created":"2026-03-12","cited_by_count":11,"area":null},{"id":"2405.19279","title":"Understanding and Minimising Outlier Features in Neural Network Training","created":"2024-05-29","cited_by_count":10,"area":null},{"id":"2508.03616","title":"Hidden Dynamics of Massive Activations in Transformer Training","created":"2025-08-05","cited_by_count":10,"area":null},{"id":"2509.09679","title":"ButterflyQuant: Ultra-low-bit LLM Quantization through Learnable Orthogonal Butterfly Transforms","created":"2025-09-11","cited_by_count":10,"area":null},{"id":"2405.06604","title":"Explaining Text Similarity in Transformer Models","created":"2024-05-10","cited_by_count":9,"area":null},{"id":"2506.12044","title":"Why Do Some Inputs Break Low-Bit LLM Quantization?","created":"2025-05-24","cited_by_count":9,"area":null},{"id":"2505.00598","title":"Fast and Low-Cost Genomic Foundation Models via Outlier Removal","created":"2025-05-01","cited_by_count":8,"area":null},{"id":"2510.09017","title":"Value-State Gated Attention for Mitigating Extreme-Token Phenomena in Transformers","created":"2025-10-10","cited_by_count":8,"area":null},{"id":"2405.13938","title":"eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization","created":"2024-05-22","cited_by_count":7,"area":null},{"id":"2412.20185","title":"DecDEC: A Systems Approach to Advancing Low-Bit LLM Quantization","created":"2024-12-28","cited_by_count":7,"area":null},{"id":"2501.19377","title":"SELMA: A Speech-Enabled Language Model for Virtual Assistant Interactions","created":"2025-01-31","cited_by_count":7,"area":null},{"id":"2504.05346","title":"Thanos: A Block-wise Pruning Algorithm for Efficient Large Language Model Compression","created":"2025-04-06","cited_by_count":7,"area":null},{"id":"2505.17101","title":"A quantitative analysis of semantic information in deep representations of text and images","created":"2025-05-21","cited_by_count":7,"area":null},{"id":"2601.04765","title":"Differential syntactic and semantic encoding in LLMs","created":"2026-01-08","cited_by_count":7,"area":null},{"id":"2310.13040","title":"Interpreting CLIP: Insights on the Robustness to ImageNet Distribution Shifts","created":"2023-10-19","cited_by_count":6,"area":null},{"id":"2509.23500","title":"Beyond Outliers: A Study of Optimizers Under Quantization","created":"2025-09-27","cited_by_count":6,"area":null},{"id":"2511.23225","title":"TWEO: Transformers Without Extreme Outliers Enables FP8 Training And Quantization For Dummies","created":"2025-11-28","cited_by_count":6,"area":null},{"id":"2602.01203","title":"Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse","created":"2026-02-01","cited_by_count":5,"area":"moe"},{"id":"2605.19660","title":"OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond","created":"2026-05-19","cited_by_count":5,"area":"inference"},{"id":"2505.21670","title":"Rethinking the Outlier Distribution in Large Language Models: An In-depth Study","created":"2025-05-27","cited_by_count":4,"area":"lr"},{"id":"2506.20752","title":"Characterization and Mitigation of Training Instabilities in Microscaling Formats","created":"2025-06-25","cited_by_count":4,"area":null},{"id":"2603.08727","title":"ARKV: Adaptive and Resource-Efficient KV Cache Management under Limited Memory Budget for Long-Context Inference in LLMs","created":"2026-02-19","cited_by_count":4,"area":"agents"},{"id":"2605.25880","title":"The Quantization Benefits of Residual-Free Transformers","created":"2026-05-25","cited_by_count":4,"area":null},{"id":"2406.11022","title":"Outlier Reduction with Gated Attention for Improved Post-training Quantization in Large Sequence-to-sequence Speech Foundation Models","created":"2024-06-16","cited_by_count":3,"area":"distill"},{"id":"2410.09637","title":"ReLU's Revival: On the Entropic Overload in Normalization-Free Large Language Models","created":"2024-10-12","cited_by_count":3,"area":null},{"id":"2410.13060","title":"AERO: Entropy-Guided Framework for Private LLM Inference","created":"2024-10-16","cited_by_count":3,"area":null},{"id":"2503.21718","title":"Not a nuisance but a useful heuristic: Outlier dimensions favor frequent tokens in language models","created":"2025-03-27","cited_by_count":3,"area":null},{"id":"2603.17771","title":"Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers","created":"2026-03-18","cited_by_count":3,"area":null},{"id":"2604.14191","title":"Attention to Mamba: A Recipe for Cross-Architecture Distillation","created":"2026-04-01","cited_by_count":3,"area":"ssm"},{"id":"2605.18898","title":"A Two-Parameter Weibull Framework for Diagnosing Transformer Weight Distributions","created":"2026-05-17","cited_by_count":3,"area":null},{"id":"2606.19005","title":"Sumi: Open Uniform Diffusion Language Model from Scratch","created":"2026-06-17","cited_by_count":3,"area":"diffusion"},{"id":"2402.12102","title":"Is It a Free Lunch for Removing Outliers during Pretraining?","created":"2024-02-19","cited_by_count":2,"area":null},{"id":"2403.06497","title":"QuantTune: Optimizing Model Quantization with Adaptive Outlier-Driven Fine Tuning","created":"2024-03-11","cited_by_count":2,"area":null},{"id":"2406.11016","title":"Optimized Speculative Sampling for GPU Hardware Accelerators","created":"2024-06-16","cited_by_count":2,"area":"inference"},{"id":"2410.11298","title":"Sorted Weight Sectioning for Energy-Efficient Unstructured Sparse DNNs on Compute-in-Memory Crossbars","created":"2024-10-15","cited_by_count":2,"area":null},{"id":"2501.15225","title":"SEAL: Scaling to Emphasize Attention for Long-Context Retrieval","created":"2025-01-25","cited_by_count":2,"area":null},{"id":"2505.22811","title":"Highly Efficient and Effective LLMs with Multi-Boolean Architectures","created":"2025-05-28","cited_by_count":2,"area":null},{"id":"2507.01532","title":"Exploring Pose-based Sign Language Translation: Ablation Studies and Attention Insights","created":"2025-07-02","cited_by_count":2,"area":null},{"id":"2510.00034","title":"Review of Hallucination Understanding in Large Language and Vision Models","created":"2025-09-26","cited_by_count":2,"area":null},{"id":"2510.19498","title":"Energy-Efficient and Dequantization-Free Q-LLMs: A Spiking Neural Network Approach to Salient Value Mitigation","created":"2025-10-22","cited_by_count":2,"area":null},{"id":"2605.08504","title":"A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models","created":"2026-05-08","cited_by_count":2,"area":null},{"id":"2605.17887","title":"Attention Sinks and Outliers in Attention Residuals","created":"2026-05-18","cited_by_count":2,"area":null},{"id":"2605.31518","title":"On the Relationship Between Activation Outliers and Feature Death in Sparse Autoencoders","created":"2026-05-29","cited_by_count":2,"area":"mechinterp"},{"id":"2606.02288","title":"Massive Spikes in LLMs are Bias Vectors: Mechanistic Uncovering and Spike-Free Quantization","created":"2026-06-01","cited_by_count":2,"area":null},{"id":"2307.13365","title":"Pay Attention to What You Need","created":"2023-07-25","cited_by_count":1,"area":null},{"id":"2308.04832","title":"TSSR: A Truncated and Signed Square Root Activation Function for Neural Networks","created":"2023-08-09","cited_by_count":1,"area":null}],"next_cursor":"100"}