{"total":273,"rows":[{"id":"2310.01801","title":"Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs","created":"2023-10-03","cited_by_count":474,"area":"inference"},{"id":"2102.04906","title":"Dynamic Neural Networks: A Survey","created":"2021-02-09","cited_by_count":408,"area":null},{"id":"2106.04554","title":"A Survey of Transformers","created":"2021-06-08","cited_by_count":401,"area":null},{"id":"2112.07658","title":"AdaViT: Adaptive Tokens for Efficient Vision Transformer","created":"2021-12-14","cited_by_count":362,"area":null},{"id":"2207.07061","title":"Confident Adaptive Language Modeling","created":"2022-07-14","cited_by_count":310,"area":"inference"},{"id":"2404.14294","title":"A Survey on Efficient Inference for Large Language Models","created":"2024-04-22","cited_by_count":205,"area":"inference"},{"id":"2204.09656","title":"A Fast Post-Training Pruning Framework for Transformers","created":"2022-03-29","cited_by_count":162,"area":null},{"id":"2402.16363","title":"LLM Inference Unveiled: Survey and Roofline Model Insights","created":"2024-02-26","cited_by_count":152,"area":"inference"},{"id":"2310.12963","title":"AutoMix: Automatically Mixing Language Models","created":"2023-10-19","cited_by_count":125,"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":"2103.04505","title":"Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges","created":"2021-03-08","cited_by_count":115,"area":"inference"},{"id":"2305.11860","title":"Let's Sample Step by Step: Adaptive-Consistency for Efficient Reasoning and Coding with LLMs","created":"2023-05-19","cited_by_count":115,"area":"lr"},{"id":"2312.15234","title":"Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems","created":"2023-12-23","cited_by_count":102,"area":"inference"},{"id":"2307.02628","title":"SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference","created":"2023-07-05","cited_by_count":89,"area":"inference"},{"id":"2401.08092","title":"A Survey of Resource-efficient LLM and Multimodal Foundation Models","created":"2024-01-16","cited_by_count":86,"area":null},{"id":"2502.08773","title":"Universal Model Routing for Efficient LLM Inference","created":"2025-02-12","cited_by_count":86,"area":"inference"},{"id":"2201.03529","title":"Head2Toe: Utilizing Intermediate Representations for Better Transfer Learning","created":"2022-01-10","cited_by_count":83,"area":null},{"id":"2307.09476","title":"Overthinking the Truth: Understanding how Language Models Process False Demonstrations","created":"2023-07-18","cited_by_count":81,"area":"lr"},{"id":"2106.05022","title":"Adaptive Inference through Early-Exit Networks: Design, Challenges and Directions","created":"2021-06-09","cited_by_count":75,"area":"inference"},{"id":"2104.08803","title":"Consistent Accelerated Inference via Confident Adaptive Transformers","created":"2021-04-18","cited_by_count":74,"area":null},{"id":"2202.08625","title":"Revisiting Over-smoothing in BERT from the Perspective of Graph","created":"2022-02-17","cited_by_count":72,"area":null},{"id":"2403.07652","title":"Harder Tasks Need More Experts: Dynamic Routing in MoE Models","created":"2024-03-12","cited_by_count":72,"area":"moe"},{"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":"2402.09997","title":"LoraRetriever: Input-Aware LoRA Retrieval and Composition for Mixed Tasks in the Wild","created":"2024-02-15","cited_by_count":68,"area":null},{"id":"2106.04570","title":"BERT Learns to Teach: Knowledge Distillation with Meta Learning","created":"2021-06-08","cited_by_count":61,"area":"distill"},{"id":"2307.02764","title":"When Does Confidence-Based Cascade Deferral Suffice?","created":"2023-07-06","cited_by_count":60,"area":null},{"id":"2210.07795","title":"EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning","created":"2022-10-14","cited_by_count":56,"area":"mllm"},{"id":"2401.15569","title":"Efficient Tuning and Inference for Large Language Models on Textual Graphs","created":"2024-01-28","cited_by_count":54,"area":"inference"},{"id":"2305.17530","title":"PuMer: Pruning and Merging Tokens for Efficient Vision Language Models","created":"2023-05-27","cited_by_count":53,"area":"mllm"},{"id":"2111.05193","title":"A Survey on Green Deep Learning","created":"2021-11-08","cited_by_count":52,"area":null},{"id":"2410.03111","title":"LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy","created":"2024-10-04","cited_by_count":51,"area":"inference"},{"id":"2202.07105","title":"A Survey on Model Compression and Acceleration for Pretrained Language Models","created":"2022-02-15","cited_by_count":45,"area":null},{"id":"2309.04255","title":"LLMCad: Fast and Scalable On-device Large Language Model Inference","created":"2023-09-08","cited_by_count":45,"area":null},{"id":"2110.07038","title":"Towards Efficient NLP: A Standard Evaluation and A Strong Baseline","created":"2021-10-13","cited_by_count":44,"area":null},{"id":"2205.03286","title":"GlobEnc: Quantifying Global Token Attribution by Incorporating the Whole Encoder Layer in Transformers","created":"2022-05-06","cited_by_count":44,"area":null},{"id":"2306.04459","title":"Uncertainty in Natural Language Processing: Sources, Quantification, and Applications","created":"2023-06-05","cited_by_count":44,"area":null},{"id":"2211.11152","title":"You Need Multiple Exiting: Dynamic Early Exiting for Accelerating Unified Vision Language Model","created":"2022-11-21","cited_by_count":43,"area":"mllm"},{"id":"2106.05409","title":"Zero Time Waste: Recycling Predictions in Early Exit Neural Networks","created":"2021-06-09","cited_by_count":39,"area":"inference"},{"id":"2109.15082","title":"Towards Efficient Post-training Quantization of Pre-trained Language Models","created":"2021-09-30","cited_by_count":39,"area":null},{"id":"2111.06061","title":"Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AI","created":"2021-11-11","cited_by_count":39,"area":null},{"id":"2109.03228","title":"Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression","created":"2021-09-07","cited_by_count":38,"area":null},{"id":"2211.09961","title":"Path Independent Equilibrium Models Can Better Exploit Test-Time Computation","created":"2022-11-18","cited_by_count":38,"area":null},{"id":"2203.06345","title":"The Principle of Diversity: Training Stronger Vision Transformers Calls for Reducing All Levels of Redundancy","created":"2022-03-12","cited_by_count":35,"area":null},{"id":"2203.01670","title":"A Simple Hash-Based Early Exiting Approach For Language Understanding and Generation","created":"2022-03-03","cited_by_count":32,"area":"inference"},{"id":"2508.02694","title":"Efficient Agents: Building Effective Agents While Reducing Cost","created":"2025-07-24","cited_by_count":32,"area":"agents"},{"id":"2308.00255","title":"LGViT: Dynamic Early Exiting for Accelerating Vision Transformer","created":"2023-08-01","cited_by_count":31,"area":"inference"},{"id":"2404.14527","title":"M\\'elange: Cost Efficient Large Language Model Serving by Exploiting GPU Heterogeneity","created":"2024-04-22","cited_by_count":31,"area":null},{"id":"2109.08406","title":"Fine-Tuned Transformers Show Clusters of Similar Representations Across Layers","created":"2021-09-17","cited_by_count":30,"area":null},{"id":"2310.13671","title":"Let's Synthesize Step by Step: Iterative Dataset Synthesis with Large Language Models by Extrapolating Errors from Small Models","created":"2023-10-20","cited_by_count":30,"area":null},{"id":"2405.20915","title":"Fast yet Safe: Early-Exiting with Risk Control","created":"2024-05-31","cited_by_count":30,"area":"inference"},{"id":"2105.13878","title":"Accelerating BERT Inference for Sequence Labeling via Early-Exit","created":"2021-05-28","cited_by_count":29,"area":"inference"},{"id":"2502.01615","title":"Large Language Models Are Human-Like Internally","created":"2025-02-03","cited_by_count":29,"area":"mechinterp"},{"id":"2208.11180","title":"Auditing Membership Leakages of Multi-Exit Networks","created":"2022-08-23","cited_by_count":27,"area":null},{"id":"2303.04185","title":"Gradient-Free Structured Pruning with Unlabeled Data","created":"2023-03-07","cited_by_count":26,"area":null},{"id":"2305.12129","title":"Lifting the Curse of Capacity Gap in Distilling Language Models","created":"2023-05-20","cited_by_count":25,"area":"moe"},{"id":"2308.14363","title":"Mobile Foundation Model as Firmware","created":"2023-08-28","cited_by_count":25,"area":null},{"id":"2411.02829","title":"CE-CoLLM: Efficient and Adaptive Large Language Models Through Cloud-Edge Collaboration","created":"2024-11-05","cited_by_count":25,"area":null},{"id":"2109.02071","title":"End-to-End Self-Debiasing Framework for Robust NLU Training","created":"2021-09-05","cited_by_count":24,"area":null},{"id":"2202.07101","title":"A Survey on Dynamic Neural Networks for Natural Language Processing","created":"2022-02-15","cited_by_count":24,"area":null},{"id":"2205.07324","title":"Transkimmer: Transformer Learns to Layer-wise Skim","created":"2022-05-15","cited_by_count":24,"area":null},{"id":"2206.00621","title":"Cross-View Language Modeling: Towards Unified Cross-Lingual Cross-Modal Pre-training","created":"2022-06-01","cited_by_count":24,"area":null},{"id":"2108.04890","title":"On the Effect of Pruning on Adversarial Robustness","created":"2021-08-10","cited_by_count":23,"area":null},{"id":"2310.03686","title":"DecoderLens: Layerwise Interpretation of Encoder-Decoder Transformers","created":"2023-10-05","cited_by_count":22,"area":null},{"id":"2101.09755","title":"RomeBERT: Robust Training of Multi-Exit BERT","created":"2021-01-24","cited_by_count":19,"area":null},{"id":"2105.13792","title":"Early Exiting with Ensemble Internal Classifiers","created":"2021-05-28","cited_by_count":19,"area":null},{"id":"2203.08991","title":"AdapLeR: Speeding up Inference by Adaptive Length Reduction","created":"2022-03-16","cited_by_count":19,"area":null},{"id":"2311.09758","title":"OrchestraLLM: Efficient Orchestration of Language Models for Dialogue State Tracking","created":"2023-11-16","cited_by_count":19,"area":null},{"id":"2112.08560","title":"Block-Skim: Efficient Question Answering for Transformer","created":"2021-12-16","cited_by_count":18,"area":null},{"id":"2205.10350","title":"Lossless Acceleration for Seq2seq Generation with Aggressive Decoding","created":"2022-05-20","cited_by_count":18,"area":null},{"id":"2305.11170","title":"Efficient Prompting via Dynamic In-Context Learning","created":"2023-05-18","cited_by_count":18,"area":null},{"id":"2305.11916","title":"F-PABEE: Flexible-patience-based Early Exiting for Single-label and Multi-label text Classification Tasks","created":"2023-05-21","cited_by_count":18,"area":null},{"id":"2206.07699","title":"Write and Paint: Generative Vision-Language Models are Unified Modal Learners","created":"2022-06-15","cited_by_count":17,"area":"mllm"},{"id":"2407.02348","title":"Agreement-Based Cascading for Efficient Inference","created":"2024-07-02","cited_by_count":16,"area":null},{"id":"2507.07996","title":"Skip a Layer or Loop it? Test-Time Depth Adaptation of Pretrained LLMs","created":"2025-07-10","cited_by_count":16,"area":null},{"id":"2508.12491","title":"Cost-Aware Contrastive Routing for LLMs","created":"2025-08-17","cited_by_count":16,"area":null},{"id":"2105.13471","title":"Inspecting the concept knowledge graph encoded by modern language models","created":"2021-05-27","cited_by_count":15,"area":null},{"id":"2310.09163","title":"Jointly-Learned Exit and Inference for a Dynamic Neural Network : JEI-DNN","created":"2023-10-13","cited_by_count":15,"area":null},{"id":"2312.05385","title":"Apparate: Rethinking Early Exits to Tame Latency-Throughput Tensions in ML Serving","created":"2023-12-08","cited_by_count":15,"area":"inference"},{"id":"2505.14733","title":"The Energy Cost of Reasoning: Analyzing Energy Usage in LLMs with Test-time Compute","created":"2025-05-20","cited_by_count":15,"area":"lr"},{"id":"2107.00175","title":"Elbert: Fast Albert with Confidence-Window Based Early Exit","created":"2021-07-01","cited_by_count":14,"area":"inference"},{"id":"2305.12228","title":"Dynamic Transformers Provide a False Sense of Efficiency","created":"2023-05-20","cited_by_count":14,"area":"inference"},{"id":"2312.08890","title":"Defenses in Adversarial Machine Learning: A Survey","created":"2023-12-13","cited_by_count":14,"area":null},{"id":"2406.16989","title":"Retrieval-Augmented Mixture of LoRA Experts for Uploadable Machine Learning","created":"2024-06-24","cited_by_count":14,"area":"rag"},{"id":"2505.17604","title":"Adaptive Semantic Token Communication for Transformer-based Edge Inference","created":"2025-05-23","cited_by_count":14,"area":null},{"id":"2101.00416","title":"Improving Sequence-to-Sequence Pre-training via Sequence Span Rewriting","created":"2021-01-02","cited_by_count":13,"area":null},{"id":"2203.14380","title":"Pyramid-BERT: Reducing Complexity via Successive Core-set based Token Selection","created":"2022-03-27","cited_by_count":13,"area":null},{"id":"2204.06328","title":"HuBERT-EE: Early Exiting HuBERT for Efficient Speech Recognition","created":"2022-04-13","cited_by_count":13,"area":"inference"},{"id":"2210.10041","title":"Hidden State Variability of Pretrained Language Models Can Guide Computation Reduction for Transfer Learning","created":"2022-10-18","cited_by_count":13,"area":null},{"id":"2309.09195","title":"SplitEE: Early Exit in Deep Neural Networks with Split Computing","created":"2023-09-17","cited_by_count":13,"area":"inference"},{"id":"2312.11882","title":"ConsistentEE: A Consistent and Hardness-Guided Early Exiting Method for Accelerating Language Models Inference","created":"2023-12-19","cited_by_count":13,"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":"2405.15039","title":"CEEBERT: Cross-Domain Inference in Early Exit BERT","created":"2024-05-23","cited_by_count":12,"area":"inference"},{"id":"2505.17616","title":"Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments","created":"2025-05-23","cited_by_count":12,"area":"agents"},{"id":"2506.06884","title":"FREE: Fast and Robust Vision Language Models with Early Exits","created":"2025-06-07","cited_by_count":12,"area":"mllm"},{"id":"2510.12773","title":"Dr.LLM: Dynamic Layer Routing in LLMs","created":"2025-10-14","cited_by_count":12,"area":null},{"id":"2607.20519","title":"Adaptive Depth in Looped Transformers: Diagnosing Learned Halting Gates and Trajectory Readouts","created":"2026-07-08","cited_by_count":12,"area":null},{"id":"2204.02426","title":"OccamNets: Mitigating Dataset Bias by Favoring Simpler Hypotheses","created":"2022-04-05","cited_by_count":11,"area":null},{"id":"2205.15237","title":"VLUE: A Multi-Task Benchmark for Evaluating Vision-Language Models","created":"2022-05-30","cited_by_count":11,"area":"mllm"},{"id":"2301.12378","title":"Towards Inference Efficient Deep Ensemble Learning","created":"2023-01-29","cited_by_count":11,"area":null},{"id":"2407.06677","title":"Mixture-of-Modules: Reinventing Transformers as Dynamic Assemblies of Modules","created":"2024-07-09","cited_by_count":11,"area":null}],"next_cursor":"100"}