{"total":236,"rows":[{"id":"2211.17192","title":"Fast Inference from Transformers via Speculative Decoding","created":"2022-11-30","cited_by_count":1653,"area":"inference"},{"id":"2006.16668","title":"GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding","created":"2020-06-30","cited_by_count":1631,"area":"moe"},{"id":"2306.13643","title":"LightGlue: Local Feature Matching at Light Speed","created":"2023-06-23","cited_by_count":590,"area":null},{"id":"2203.14680","title":"Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space","created":"2022-03-28","cited_by_count":532,"area":null},{"id":"2102.04906","title":"Dynamic Neural Networks: A Survey","created":"2021-02-09","cited_by_count":408,"area":null},{"id":"2309.03883","title":"DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models","created":"2023-09-07","cited_by_count":376,"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":"2404.16710","title":"LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding","created":"2024-04-25","cited_by_count":307,"area":"inference"},{"id":"2404.02258","title":"Mixture-of-Depths: Dynamically allocating compute in transformer-based language models","created":"2024-04-02","cited_by_count":258,"area":null},{"id":"2302.07863","title":"Speculative Decoding with Big Little Decoder","created":"2023-02-15","cited_by_count":164,"area":"inference"},{"id":"2507.10524","title":"Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation","created":"2025-07-14","cited_by_count":133,"area":null},{"id":"2411.02359","title":"DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution","created":"2024-11-04","cited_by_count":117,"area":"lr"},{"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":"2110.03742","title":"Beyond Distillation: Task-level Mixture-of-Experts for Efficient Inference","created":"2021-09-24","cited_by_count":96,"area":"moe"},{"id":"2403.02181","title":"Not All Layers of LLMs Are Necessary During Inference","created":"2024-03-04","cited_by_count":96,"area":null},{"id":"2410.11779","title":"MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation","created":"2024-10-15","cited_by_count":88,"area":"mllm"},{"id":"2310.05424","title":"Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding","created":"2023-10-09","cited_by_count":80,"area":"inference"},{"id":"2209.00099","title":"Efficient Methods for Natural Language Processing: A Survey","created":"2022-08-31","cited_by_count":78,"area":null},{"id":"2410.20672","title":"Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA","created":"2024-10-28","cited_by_count":78,"area":null},{"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":"2103.14636","title":"A Practical Survey on Faster and Lighter Transformers","created":"2021-03-26","cited_by_count":73,"area":null},{"id":"2410.03577","title":"Look Twice Before You Answer: Memory-Space Visual Retracing for Hallucination Mitigation in Multimodal Large Language Models","created":"2024-10-04","cited_by_count":72,"area":"mllm"},{"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":"2305.17328","title":"Zero-TPrune: Zero-Shot Token Pruning through Leveraging of the Attention Graph in Pre-Trained Transformers","created":"2023-05-27","cited_by_count":62,"area":"inference"},{"id":"2602.11451","title":"LoopFormer: Elastic-Depth Looped Transformers for Latent Reasoning via Shortcut Modulation","created":"2026-02-11","cited_by_count":61,"area":"lr"},{"id":"2307.02764","title":"When Does Confidence-Based Cascade Deferral Suffice?","created":"2023-07-06","cited_by_count":60,"area":null},{"id":"2604.12946","title":"Parcae: Scaling Laws For Stable Looped Language Models","created":"2026-04-14","cited_by_count":55,"area":null},{"id":"2111.05193","title":"A Survey on Green Deep Learning","created":"2021-11-08","cited_by_count":52,"area":null},{"id":"2511.07384","title":"Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence","created":"2025-11-10","cited_by_count":47,"area":"lr"},{"id":"2402.05964","title":"A Survey on Transformer Compression","created":"2024-02-05","cited_by_count":46,"area":null},{"id":"2412.01199","title":"TinyFusion: Diffusion Transformers Learned Shallow","created":"2024-12-02","cited_by_count":45,"area":"diffusion"},{"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":"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":"2203.13240","title":"Token Dropping for Efficient BERT Pretraining","created":"2022-03-24","cited_by_count":42,"area":"inference"},{"id":"2110.07515","title":"Non-Autoregressive Translation with Layer-Wise Prediction and Deep Supervision","created":"2021-10-14","cited_by_count":40,"area":null},{"id":"2407.01178","title":"$\\text{Memory}^3$: Language Modeling with Explicit Memory","created":"2024-07-01","cited_by_count":40,"area":"rag"},{"id":"2109.12575","title":"Paradigm Shift in Natural Language Processing","created":"2021-09-26","cited_by_count":39,"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":"2405.19261","title":"Faster Cascades via Speculative Decoding","created":"2024-05-29","cited_by_count":38,"area":"inference"},{"id":"2407.15018","title":"Answer, Assemble, Ace: Understanding How LMs Answer Multiple Choice Questions","created":"2024-07-21","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":"2310.07096","title":"Sparse Universal Transformer","created":"2023-10-11","cited_by_count":32,"area":"moe"},{"id":"2406.03853","title":"Speculative Decoding via Early-exiting for Faster LLM Inference with Thompson Sampling Control Mechanism","created":"2024-06-06","cited_by_count":32,"area":"inference"},{"id":"2503.10084","title":"Why Prompt Design Matters and Works: A Complexity Analysis of Prompt Search Space in LLMs","created":"2025-03-13","cited_by_count":31,"area":null},{"id":"2305.12132","title":"Can Public Large Language Models Help Private Cross-device Federated Learning?","created":"2023-05-20","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":"2305.01628","title":"The Benefits of Bad Advice: Autocontrastive Decoding across Model Layers","created":"2023-05-02","cited_by_count":29,"area":null},{"id":"2410.20771","title":"MrT5: Dynamic Token Merging for Efficient Byte-level Language Models","created":"2024-10-28","cited_by_count":28,"area":"inference"},{"id":"2106.10234","title":"Dual-view Molecule Pre-training","created":"2021-06-17","cited_by_count":27,"area":null},{"id":"2310.10845","title":"CoTFormer: A Chain-of-Thought Driven Architecture with Budget-Adaptive Computation Cost at Inference","created":"2023-10-16","cited_by_count":27,"area":"lr"},{"id":"2505.18962","title":"System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts","created":"2025-05-25","cited_by_count":27,"area":"lr"},{"id":"2105.09121","title":"Single-Layer Vision Transformers for More Accurate Early Exits with Less Overhead","created":"2021-05-19","cited_by_count":26,"area":"inference"},{"id":"2402.13720","title":"Ouroboros: Generating Longer Drafts Phrase by Phrase for Faster Speculative Decoding","created":"2024-02-21","cited_by_count":25,"area":"inference"},{"id":"2407.19985","title":"Mixture of Nested Experts: Adaptive Processing of Visual Tokens","created":"2024-07-29","cited_by_count":25,"area":"moe"},{"id":"2410.18860","title":"DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations","created":"2024-10-24","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":"2202.07101","title":"A Survey on Dynamic Neural Networks for Natural Language Processing","created":"2022-02-15","cited_by_count":24,"area":null},{"id":"2505.20353","title":"FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximation","created":"2025-05-26","cited_by_count":24,"area":"diffusion"},{"id":"2509.26522","title":"Entropy After </Think> for reasoning model early exiting","created":"2025-09-30","cited_by_count":24,"area":"lr"},{"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":"2404.03865","title":"FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping","created":"2024-04-05","cited_by_count":23,"area":"inference"},{"id":"2602.13517","title":"Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens","created":"2026-02-13","cited_by_count":22,"area":"lr"},{"id":"2605.19376","title":"Generative Recursive Reasoning","created":"2026-05-19","cited_by_count":20,"area":null},{"id":"2105.13792","title":"Early Exiting with Ensemble Internal Classifiers","created":"2021-05-28","cited_by_count":19,"area":null},{"id":"2311.15436","title":"Learning to Skip for Language Modeling","created":"2023-11-26","cited_by_count":19,"area":null},{"id":"2312.05662","title":"Understanding the Effect of Model Compression on Social Bias in Large Language Models","created":"2023-12-09","cited_by_count":19,"area":null},{"id":"2609.01343","title":"SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers","created":"2026-09-01","cited_by_count":18,"area":"moe"},{"id":"2404.06954","title":"Accelerating Inference in Large Language Models with a Unified Layer Skipping Strategy","created":"2024-04-10","cited_by_count":17,"area":null},{"id":"2510.18871","title":"How Do LLMs Use Their Depth?","created":"2025-10-21","cited_by_count":16,"area":null},{"id":"2106.10820","title":"Stateful ODE-Nets using Basis Function Expansions","created":"2021-06-21","cited_by_count":14,"area":null},{"id":"2407.11798","title":"PipeInfer: Accelerating LLM Inference using Asynchronous Pipelined Speculation","created":"2024-07-16","cited_by_count":14,"area":"inference"},{"id":"2305.03796","title":"Transformer Working Memory Enables Regular Language Reasoning and Natural Language Length Extrapolation","created":"2023-05-05","cited_by_count":13,"area":null},{"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":"2406.16635","title":"ShadowLLM: Predictor-based Contextual Sparsity for Large Language Models","created":"2024-06-24","cited_by_count":13,"area":null},{"id":"2405.15039","title":"CEEBERT: Cross-Domain Inference in Early Exit BERT","created":"2024-05-23","cited_by_count":12,"area":"inference"},{"id":"2406.01698","title":"Demystifying AI Platform Design for Distributed Inference of Next-Generation LLM models","created":"2024-06-03","cited_by_count":12,"area":"inference"},{"id":"2602.11698","title":"SpiralFormer: Looped Transformers Can Learn Hierarchical Dependencies via Multi-Resolution Recursion","created":"2026-02-12","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":"2407.06677","title":"Mixture-of-Modules: Reinventing Transformers as Dynamic Assemblies of Modules","created":"2024-07-09","cited_by_count":11,"area":null},{"id":"2409.14199","title":"Loop Neural Networks for Parameter Sharing","created":"2024-09-21","cited_by_count":11,"area":null},{"id":"2503.01275","title":"Enhancing Non-English Capabilities of English-Centric Large Language Models through Deep Supervision Fine-Tuning","created":"2025-03-03","cited_by_count":11,"area":null},{"id":"2209.09480","title":"Unsupervised Early Exit in DNNs with Multiple Exits","created":"2022-09-20","cited_by_count":10,"area":null},{"id":"2210.13304","title":"ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation","created":"2022-10-24","cited_by_count":10,"area":"inference"},{"id":"2307.02092","title":"Make A Long Image Short: Adaptive Token Length for Vision Transformers","created":"2023-07-05","cited_by_count":10,"area":null},{"id":"2310.13682","title":"Optimizing Retrieval-augmented Reader Models via Token Elimination","created":"2023-10-20","cited_by_count":10,"area":"rag"},{"id":"2504.02862","title":"Towards Understanding How Knowledge Evolves in Large Vision-Language Models","created":"2025-03-31","cited_by_count":10,"area":"mllm"},{"id":"2506.21103","title":"Learning to Skip the Middle Layers of Transformers","created":"2025-06-26","cited_by_count":10,"area":"moe"},{"id":"2510.03136","title":"Beyond the Final Layer: Intermediate Representations for Better Multilingual Calibration in Large Language Models","created":"2025-10-03","cited_by_count":10,"area":null},{"id":"2510.13251","title":"Map the Flow: Revealing Hidden Pathways of Information in VideoLLMs","created":"2025-10-15","cited_by_count":10,"area":"mllm"},{"id":"2109.11745","title":"DACT-BERT: Differentiable Adaptive Computation Time for an Efficient BERT Inference","created":"2021-09-24","cited_by_count":9,"area":null},{"id":"2301.07099","title":"Adaptive Deep Neural Network Inference Optimization with EENet","created":"2023-01-15","cited_by_count":9,"area":"inference"},{"id":"2405.14105","title":"Distributed Speculative Inference (DSI): Speculation Parallelism for Provably Faster Lossless Language Model Inference","created":"2024-05-23","cited_by_count":9,"area":"inference"},{"id":"2506.18945","title":"Chain-of-Experts: Unlocking the Communication Power of Mixture-of-Experts Models","created":"2025-06-23","cited_by_count":9,"area":"moe"},{"id":"2110.10305","title":"When in Doubt, Summon the Titans: Efficient Inference with Large Models","created":"2021-10-19","cited_by_count":8,"area":null},{"id":"2502.13329","title":"Language Models Can Predict Their Own Behavior","created":"2025-02-18","cited_by_count":8,"area":"lr"},{"id":"2503.13794","title":"LED: LLM Enhanced Open-Vocabulary Object Detection without Human Curated Data Generation","created":"2025-03-18","cited_by_count":8,"area":null},{"id":"2510.07739","title":"MeSH: Memory-as-State-Highways for Recursive Transformers","created":"2025-10-09","cited_by_count":8,"area":null}],"next_cursor":"100"}