{"total":226,"rows":[{"id":"2407.14679","title":"Compact Language Models via Pruning and Knowledge Distillation","created":"2024-07-19","cited_by_count":148,"area":"distill"},{"id":"2406.19384","title":"The Remarkable Robustness of LLMs: Stages of Inference?","created":"2024-06-27","cited_by_count":146,"area":null},{"id":"2405.14366","title":"MiniCache: KV Cache Compression in Depth Dimension for Large Language Models","created":"2024-05-23","cited_by_count":125,"area":"inference"},{"id":"2506.05218","title":"MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigm","created":"2025-06-05","cited_by_count":87,"area":null},{"id":"2408.11796","title":"LLM Pruning and Distillation in Practice: The Minitron Approach","created":"2024-08-21","cited_by_count":82,"area":null},{"id":"2507.06203","title":"A Survey on Latent Reasoning","created":"2025-07-08","cited_by_count":81,"area":"lr"},{"id":"2411.04986","title":"The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities","created":"2024-11-07","cited_by_count":69,"area":null},{"id":"2411.17491","title":"What's in the Image? A Deep-Dive into the Vision of Vision Language Models","created":"2024-11-26","cited_by_count":68,"area":"mllm"},{"id":"2603.15031","title":"Attention Residuals","created":"2026-03-16","cited_by_count":65,"area":null},{"id":"2502.05795","title":"The Curse of Depth in Large Language Models","created":"2025-02-09","cited_by_count":62,"area":null},{"id":"2405.15471","title":"Emergence of a High-Dimensional Abstraction Phase in Language Transformers","created":"2024-05-24","cited_by_count":56,"area":null},{"id":"2505.13898","title":"Do Language Models Use Their Depth Efficiently?","created":"2025-05-20","cited_by_count":50,"area":null},{"id":"2412.13795","title":"Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN","created":"2024-12-18","cited_by_count":37,"area":null},{"id":"2406.14479","title":"Tracing Representation Progression: Analyzing and Enhancing Layer-Wise Similarity","created":"2024-06-20","cited_by_count":36,"area":null},{"id":"2410.10912","title":"AlphaPruning: Using Heavy-Tailed Self Regularization Theory for Improved Layer-wise Pruning of Large Language Models","created":"2024-10-14","cited_by_count":36,"area":null},{"id":"2407.09141","title":"Accuracy is Not All You Need","created":"2024-07-12","cited_by_count":31,"area":null},{"id":"2502.12170","title":"MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections","created":"2025-02-13","cited_by_count":31,"area":null},{"id":"2406.05981","title":"ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization","created":"2024-06-10","cited_by_count":30,"area":null},{"id":"2510.07358","title":"Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts","created":"2025-10-08","cited_by_count":24,"area":"lr"},{"id":"2405.16700","title":"Implicit Multimodal Alignment: On the Generalization of Frozen LLMs to Multimodal Inputs","created":"2024-05-26","cited_by_count":23,"area":null},{"id":"2502.06857","title":"Gemstones: A Model Suite for Multi-Faceted Scaling Laws","created":"2025-02-07","cited_by_count":20,"area":null},{"id":"2406.16330","title":"Pruning via Merging: Compressing LLMs via Manifold Alignment Based Layer Merging","created":"2024-06-24","cited_by_count":19,"area":null},{"id":"2410.03960","title":"SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation","created":"2024-10-04","cited_by_count":19,"area":"inference"},{"id":"2410.17875","title":"Understanding Layer Significance in LLM Alignment","created":"2024-10-23","cited_by_count":19,"area":"alignment"},{"id":"2405.17890","title":"SLMRec: Distilling Large Language Models into Small for Sequential Recommendation","created":"2024-05-28","cited_by_count":18,"area":"distill"},{"id":"2502.07780","title":"DarwinLM: Evolutionary Structured Pruning of Large Language Models","created":"2025-02-11","cited_by_count":18,"area":null},{"id":"2502.14377","title":"RelaCtrl: Relevance-Guided Efficient Control for Diffusion Transformers","created":"2025-02-20","cited_by_count":18,"area":"diffusion"},{"id":"2410.17881","title":"AdaRankGrad: Adaptive Gradient-Rank and Moments for Memory-Efficient LLMs Training and Fine-Tuning","created":"2024-10-23","cited_by_count":17,"area":null},{"id":"2411.19146","title":"Puzzle: Distillation-Based NAS for Inference-Optimized LLMs","created":"2024-11-28","cited_by_count":17,"area":null},{"id":"2406.17415","title":"Layer-Wise Quantization: A Pragmatic and Effective Method for Quantizing LLMs Beyond Integer Bit-Levels","created":"2024-06-25","cited_by_count":16,"area":null},{"id":"2412.14590","title":"MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design","created":"2024-12-19","cited_by_count":16,"area":null},{"id":"2503.18893","title":"xKV: Cross-Layer KV-Cache Compression via Aligned Singular Vector Extraction","created":"2025-03-24","cited_by_count":16,"area":"inference"},{"id":"2510.18871","title":"How Do LLMs Use Their Depth?","created":"2025-10-21","cited_by_count":16,"area":null},{"id":"2510.02091","title":"Demystifying the Roles of LLM Layers in Retrieval, Knowledge, and Reasoning","created":"2025-10-02","cited_by_count":15,"area":"lr"},{"id":"2512.20607","title":"Saddle-to-Saddle Dynamics Explains A Simplicity Bias Across Neural Network Architectures","created":"2025-12-23","cited_by_count":15,"area":null},{"id":"2408.04532","title":"How Transformers Utilize Multi-Head Attention in In-Context Learning? A Case Study on Sparse Linear Regression","created":"2024-08-08","cited_by_count":14,"area":null},{"id":"2505.02819","title":"ReplaceMe: Network Simplification via Depth Pruning and Transformer Block Linearization","created":"2025-05-05","cited_by_count":13,"area":null},{"id":"2505.17646","title":"Unveiling the Basin-Like Loss Landscape in Large Language Models","created":"2025-05-23","cited_by_count":13,"area":null},{"id":"2507.00390","title":"MoNE: Replacing Redundant Experts with Lightweight Novices for Structured Pruning of MoE","created":"2025-07-01","cited_by_count":13,"area":"moe"},{"id":"2410.11042","title":"Persistent Topological Features in Large Language Models","created":"2024-10-14","cited_by_count":12,"area":null},{"id":"2411.10606","title":"AmoebaLLM: Constructing Any-Shape Large Language Models for Efficient and Instant Deployment","created":"2024-11-15","cited_by_count":12,"area":null},{"id":"2501.17771","title":"2SSP: A Two-Stage Framework for Structured Pruning of LLMs","created":"2025-01-29","cited_by_count":12,"area":null},{"id":"2505.23811","title":"LayerIF: Estimating Layer Quality for Large Language Models using Influence Functions","created":"2025-05-27","cited_by_count":12,"area":null},{"id":"2409.13199","title":"CFSP: An Efficient Structured Pruning Framework for LLMs with Coarse-to-Fine Activation Information","created":"2024-09-20","cited_by_count":11,"area":null},{"id":"2506.04179","title":"SkipGPT: Dynamic Layer Pruning Reinvented with Token Awareness and Module Decoupling","created":"2025-06-04","cited_by_count":11,"area":null},{"id":"2507.08064","title":"PUMA: Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning","created":"2025-07-10","cited_by_count":11,"area":"distill"},{"id":"2602.17951","title":"ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Models","created":"2026-02-20","cited_by_count":11,"area":"vla"},{"id":"2410.13184","title":"Router-Tuning: A Simple and Effective Approach for Enabling Dynamic-Depth in Transformers","created":"2024-10-17","cited_by_count":10,"area":null},{"id":"2506.21103","title":"Learning to Skip the Middle Layers of Transformers","created":"2025-06-26","cited_by_count":10,"area":"moe"},{"id":"2602.08064","title":"SiameseNorm: Breaking the Barrier to Reconciling Pre/Post-Norm","created":"2026-02-08","cited_by_count":10,"area":null},{"id":"2605.06510","title":"Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models","created":"2026-05-07","cited_by_count":10,"area":null},{"id":"2411.03513","title":"Change Is the Only Constant: Dynamic LLM Slicing based on Layer Redundancy","created":"2024-11-05","cited_by_count":9,"area":null},{"id":"2505.03531","title":"Faster MoE LLM Inference for Extremely Large Models","created":"2025-05-06","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":"2510.12581","title":"LayerSync: Self-aligning Intermediate Layers","created":"2025-10-14","cited_by_count":9,"area":"diffusion"},{"id":"2510.22026","title":"Normalization in Attention Dynamics","created":"2025-10-24","cited_by_count":9,"area":null},{"id":"2511.20223","title":"V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMs","created":"2025-11-25","cited_by_count":9,"area":"mllm"},{"id":"2602.05970","title":"Inverse Depth Scaling From Most Layers Being Similar","created":"2026-02-05","cited_by_count":9,"area":null},{"id":"2410.01028","title":"Draft on the Fly: Adaptive Self-Speculative Decoding using Cosine Similarity","created":"2024-10-01","cited_by_count":8,"area":"inference"},{"id":"2410.09982","title":"Self-Data Distillation for Recovering Quality in Pruned Large Language Models","created":"2024-10-13","cited_by_count":8,"area":null},{"id":"2504.10724","title":"HELIOS: Adaptive Model And Early-Exit Selection for Efficient LLM Inference Serving","created":"2025-04-14","cited_by_count":8,"area":"inference"},{"id":"2505.01731","title":"Efficient Shapley Value-based Non-Uniform Pruning of Large Language Models","created":"2025-05-03","cited_by_count":8,"area":null},{"id":"2510.00345","title":"Cutting the Skip: Training Residual-Free Transformers","created":"2025-09-30","cited_by_count":8,"area":null},{"id":"2601.06787","title":"Garbage Attention in Large Language Models: BOS Sink Heads and Sink-aware Pruning","created":"2026-01-11","cited_by_count":8,"area":null},{"id":"2607.13124","title":"ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation","created":"2026-07-14","cited_by_count":8,"area":"distill"},{"id":"2406.14553","title":"xCOMET-lite: Bridging the Gap Between Efficiency and Quality in Learned MT Evaluation Metrics","created":"2024-06-20","cited_by_count":7,"area":null},{"id":"2410.09454","title":"Skipping Computations in Multimodal LLMs","created":"2024-10-12","cited_by_count":7,"area":"mllm"},{"id":"2412.14628","title":"Qua$^2$SeDiMo: Quantifiable Quantization Sensitivity of Diffusion Models","created":"2024-12-19","cited_by_count":7,"area":"diffusion"},{"id":"2502.02790","title":"Leveraging the true depth of LLMs","created":"2025-02-05","cited_by_count":7,"area":null},{"id":"2507.23362","title":"Short-LVLM: Compressing and Accelerating Large Vision-Language Models by Pruning Redundant Layers","created":"2025-07-31","cited_by_count":7,"area":"mllm"},{"id":"2508.02215","title":"LeanK: Learnable K Cache Channel Pruning for Efficient Decoding","created":"2025-08-04","cited_by_count":7,"area":null},{"id":"2410.14578","title":"Large Language Models Are Overparameterized Text Encoders","created":"2024-10-18","cited_by_count":6,"area":null},{"id":"2505.23807","title":"DLP: Dynamic Layerwise Pruning in Large Language Models","created":"2025-05-27","cited_by_count":6,"area":null},{"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":"2412.13435","title":"Lightweight Safety Classification Using Pruned Language Models","created":"2024-12-18","cited_by_count":5,"area":"safety"},{"id":"2501.02342","title":"Optimizing Small Language Models for In-Vehicle Function-Calling","created":"2025-01-04","cited_by_count":5,"area":"agents"},{"id":"2502.06766","title":"Exploiting Sparsity for Long Context Inference: Million Token Contexts on Commodity GPUs","created":"2025-02-10","cited_by_count":5,"area":null},{"id":"2502.15051","title":"Approximating Latent Manifolds in Neural Networks via Vanishing Ideals","created":"2025-02-20","cited_by_count":5,"area":null},{"id":"2504.21174","title":"Efficient LLMs with AMP: Attention Heads and MLP Pruning","created":"2025-04-29","cited_by_count":5,"area":null},{"id":"2507.18212","title":"Prune&Comp: Free Lunch for Layer-Pruned LLMs via Iterative Pruning with Magnitude Compensation","created":"2025-07-24","cited_by_count":5,"area":null},{"id":"2606.25008","title":"Neural Scaling Universality: If Exponents Are Fixed, Time to Understand Coefficients","created":"2026-06-23","cited_by_count":5,"area":null},{"id":"2406.03768","title":"Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical Perspective","created":"2024-06-06","cited_by_count":4,"area":null},{"id":"2406.08298","title":"AdaNCA: Neural Cellular Automata As Adaptors For More Robust Vision Transformer","created":"2024-06-12","cited_by_count":4,"area":null},{"id":"2411.01179","title":"Hollowed Net for On-Device Personalization of Text-to-Image Diffusion Models","created":"2024-11-02","cited_by_count":4,"area":"diffusion"},{"id":"2412.13757","title":"Federated Source-free Domain Adaptation for Classification: Weighted Cluster Aggregation for Unlabeled Data","created":"2024-12-18","cited_by_count":4,"area":null},{"id":"2502.08690","title":"Skrr: Skip and Re-use Text Encoder Layers for Memory Efficient Text-to-Image Generation","created":"2025-02-12","cited_by_count":4,"area":"diffusion"},{"id":"2505.24680","title":"A Simple Linear Patch Revives Layer-Pruned Large Language Models","created":"2025-05-30","cited_by_count":4,"area":null},{"id":"2506.09714","title":"Auto-Compressing Networks","created":"2025-06-11","cited_by_count":4,"area":null},{"id":"2508.05239","title":"Pruning Large Language Models by Identifying and Preserving Functional Networks","created":"2025-08-07","cited_by_count":4,"area":null},{"id":"2512.02351","title":"Understanding and Harnessing Sparsity in Unified Multimodal Models","created":"2025-12-02","cited_by_count":4,"area":"mllm"},{"id":"2603.15389","title":"When Does Sparsity Mitigate the Curse of Depth in LLMs","created":"2026-03-16","cited_by_count":4,"area":null},{"id":"2604.03258","title":"SoLA: Leveraging Soft Activation Sparsity and Low-Rank Decomposition for Large Language Model Compression","created":"2026-03-12","cited_by_count":4,"area":null},{"id":"2605.14075","title":"Rethinking Layer Relevance in Large Language Models Beyond Cosine Similarity","created":"2026-05-13","cited_by_count":4,"area":"lr"},{"id":"2606.03465","title":"Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression","created":"2026-06-02","cited_by_count":4,"area":"lr"},{"id":"2608.10837","title":"TACTICL: Task-Aware Compression of Tabular ICL Models","created":"2026-08-11","cited_by_count":4,"area":null},{"id":"2409.15342","title":"Recall: Empowering Multimodal Embedding for Edge Devices","created":"2024-09-09","cited_by_count":3,"area":null},{"id":"2502.15796","title":"Pruning as a Defense: Reducing Memorization in Large Language Models","created":"2025-02-18","cited_by_count":3,"area":null},{"id":"2504.15120","title":"Kuwain 1.5B: An Arabic SLM via Language Injection","created":"2025-04-21","cited_by_count":3,"area":null},{"id":"2504.20922","title":"DYNAMAX: Dynamic computing for Transformers and Mamba based architectures","created":"2025-04-29","cited_by_count":3,"area":"ssm"},{"id":"2506.20480","title":"GPTailor: Large Language Model Pruning Through Layer Cutting and Stitching","created":"2025-06-25","cited_by_count":3,"area":null}],"next_cursor":"100"}