长上下文论文趋势 2023–2026

主要结论

• 长上下文论文从 2023 年的 66 篇增长到 2026 年的 341 篇，三年间扩大到约 5.2 倍。
• NeurIPS 是发长上下文论文最多的会议，2023–2026 年共 298 篇，占四个会议总量的 36%。
• 交叉方向里涨得最快的是AI 智能体：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 45 篇。
• 在「长上下文与记忆」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的检索记忆是 6.6 倍，长上下文是 5.2 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026Accelerating Long-Context LLM Prefill via Layer-wise Progressive Token Pruning in Local Deployment
• ICML 2026AdaSplash-2: Faster Differentiable Sparse Attention
• ICLR 2026An evolutionary perspective on modes of learning in Transformers
• CVPR 2026AR2-4FV Anchored Referring and Re-identification for Long-Term Grounding in Fixed-View
• NeurIPS 2025A Controllable Examination for Long-Context Language Models
• ICML 2025CateKV: On Sequential Consistency for Long-Context LLM Inference Acceleration
• ICLR 2025$\text{D}_{2}\text{O}$: Dynamic Discriminative Operations for Efficient Long-Context Inference of Large Language Models
• CVPR 2025BIMBA: Selective-Scan Compression for Long-Range Video Question Answering
• NeurIPS 2024[Re] CUDA: Curriculum of Data Augmentation for Long‐tailed Recognition
• ICML 2024Benchmarking and Building Long-Context Retrieval Models with LoCo and M2-BERT
• ICLR 2024CLEX: Continuous Length Extrapolation for Large Language Models
• CVPR 2024GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes
• NeurIPS 2023A Single 2D Pose with Context is Worth Hundreds for 3D Human Pose Estimation
• ICML 2023CAB: Comprehensive Attention Benchmarking on Long Sequence Modeling
• ICLR 2023$k$NN Prompting: Beyond-Context Learning with Calibration-Free Nearest Neighbor Inference
• CVPR 2023AutoAD: Movie Description in Context
• NeurIPS 2026AdaCal: Adaptive Calibration for Robust Sparse Attention in Long-Context LLMs
• ICML 2026Artificial Hippocampus Networks for Efficient Long-Context Modeling
• ICLR 2026AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism
• CVPR 2026AstraNav-Memory Contexts Compression for Long Memory

全部长上下文论文

共 130 篇，第 3 / 3 页

101. NeurIPS 2026SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking
102. NeurIPS 2026Short-Context Dominance: How Much Local Context Natural Language Actually Needs?
103. NeurIPS 2026SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks
104. NeurIPS 2026SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining
105. NeurIPS 2026SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference
106. NeurIPS 2026SpectralKV: Redundancy-Aware KV Cache Compression via Spectral Coreset Selection
107. NeurIPS 2026SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics
108. NeurIPS 2026SurvivalPFN: Amortizing Survival Prediction via In-Context Bayesian Inference
109. NeurIPS 2026SWE-Marathon: Can AI Agents Autonomously Complete Ultra-Long-Horizon Software Work?
110. NeurIPS 2026Test time training enhances in-context learning of nonlinear functions
111. NeurIPS 2026The $1/\mathcal{W}$ Law: Context Length is the Dominant Energy Lever in LLM Inference Fleets
112. NeurIPS 2026The Key to Going Linear: Analysis-Driven Transformer Linearization
113. NeurIPS 2026The Long-Run Distribution of Regularized Learning in Non-Concave Games: A Large Deviations Approach
114. NeurIPS 2026The PokeAgent Challenge: Competitive and Long-Context Learning at Scale
115. NeurIPS 2026The Price of Choosing Examples: Context Overfitting in Adaptive In-Context Learning
116. NeurIPS 2026To Copy or Not to Copy: Controlling Speculative Decoding via Intrinsic Model Signals
117. NeurIPS 2026Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context
118. NeurIPS 2026Transformers Can Learn Multiclass Classification In-Context: Isotropy Governs Generalization
119. NeurIPS 2026Trimming the Long-Tail of Visual World Modeling Evaluation
120. NeurIPS 2026TSFM Meets LLM: Context as Covariate
121. NeurIPS 2026TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting
122. NeurIPS 2026UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification
123. NeurIPS 2026UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
124. NeurIPS 2026VINCIE-NExT: Unlocking Video Editing from Images via In-Context Modeling
125. NeurIPS 2026Virtual Head Attention
126. NeurIPS 2026VL-DocIR: A Benchmark for Vision-Based Long Document Retrieval
127. NeurIPS 2026Volatility-Whitened Probabilistic Residual Modeling for Long-Term Time Series Forecasting
128. NeurIPS 2026What Remains in Sight? Autoregressive Video Decoding as Representation-Guided Context Rewriting
129. NeurIPS 2026Why Do Time Series Models Need Long Context Windows?
130. NeurIPS 2026Why Transformers Struggle with Distribution-Independent In-Context Learning

常见问题

2023–2026 年四大会议一共有多少篇长上下文论文？

共 838 篇，按年份分别是 2023 年 66 篇、2024 年 176 篇、2025 年 255 篇、2026 年 341 篇。

NeurIPS 2026 有多少篇长上下文论文？

NeurIPS 2026 有 130 篇长上下文论文，上一届是 98 篇。

ICLR 2026 有多少篇长上下文论文？

ICLR 2026 有 82 篇长上下文论文，上一届是 87 篇。

ICML 2026 有多少篇长上下文论文？

ICML 2026 有 98 篇长上下文论文，上一届是 53 篇。

CVPR 2026 有多少篇长上下文论文？

CVPR 2026 有 31 篇长上下文论文，上一届是 17 篇。

哪个会议的长上下文论文最多？

NeurIPS，2023–2026 年共 298 篇。