长上下文论文趋势 2023–2026
长窗口建模
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 838 篇被归入该主题的论文。
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。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 18 | 18 | 15 | 15 | 66 |
| 2024 | 52 | 43 | 60 | 21 | 176 |
| 2025 | 98 | 87 | 53 | 17 | 255 |
| 2026 | 130 | 82 | 98 | 31 | 341 |
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 341 篇，第 7 / 7 页
- NeurIPS 2026Personalize-then-Store: Benchmarking and Learning Personalized Memory for Long-Horizon Agents
- NeurIPS 2026Point4D: Long-range 4D Motion Reconstruction
- NeurIPS 2026Preventing Error Cascades in Long-Horizon Multimodal Agents with Edge-Reliability Graph Memory
- NeurIPS 2026PRISM: Primitive Routing via In-context Skill Mixing for Lifelong VLA
- NeurIPS 2026Pure Exploration Beyond Reward Feedback: The Role of Post-Action Context
- NeurIPS 2026Q-Focus: Let the Question Guide What to See in Long Videos
- NeurIPS 2026Recursive Language Models
- NeurIPS 2026Resilient Latent Readouts for Long-Context Question Answering
- NeurIPS 2026RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades
- NeurIPS 2026ROLLVERIFY: BRIDGING EFFICIENCY AND ACCURACY IN LONG-TAIL ROLLOUT REINFORCEMENT LEARNING
- NeurIPS 2026SaaSBench: Exploring the Boundaries of Coding Agents in Long-Horizon Enterprise SaaS Engineering
- NeurIPS 2026SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking
- NeurIPS 2026Short-Context Dominance: How Much Local Context Natural Language Actually Needs?
- NeurIPS 2026SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks
- NeurIPS 2026SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining
- NeurIPS 2026SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference
- NeurIPS 2026SpectralKV: Redundancy-Aware KV Cache Compression via Spectral Coreset Selection
- NeurIPS 2026SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics
- NeurIPS 2026SurvivalPFN: Amortizing Survival Prediction via In-Context Bayesian Inference
- NeurIPS 2026SWE-Marathon: Can AI Agents Autonomously Complete Ultra-Long-Horizon Software Work?
- NeurIPS 2026Test time training enhances in-context learning of nonlinear functions
- NeurIPS 2026The $1/\mathcal{W}$ Law: Context Length is the Dominant Energy Lever in LLM Inference Fleets
- NeurIPS 2026The Key to Going Linear: Analysis-Driven Transformer Linearization
- NeurIPS 2026The Long-Run Distribution of Regularized Learning in Non-Concave Games: A Large Deviations Approach
- NeurIPS 2026The PokeAgent Challenge: Competitive and Long-Context Learning at Scale
- NeurIPS 2026The Price of Choosing Examples: Context Overfitting in Adaptive In-Context Learning
- NeurIPS 2026To Copy or Not to Copy: Controlling Speculative Decoding via Intrinsic Model Signals
- NeurIPS 2026Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context
- NeurIPS 2026Transformers Can Learn Multiclass Classification In-Context: Isotropy Governs Generalization
- NeurIPS 2026Trimming the Long-Tail of Visual World Modeling Evaluation
- NeurIPS 2026TSFM Meets LLM: Context as Covariate
- NeurIPS 2026TStruct: Learning Shared Temporal Structures for Long-Term Time-Series Forecasting
- NeurIPS 2026UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification
- NeurIPS 2026UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
- NeurIPS 2026VINCIE-NExT: Unlocking Video Editing from Images via In-Context Modeling
- NeurIPS 2026Virtual Head Attention
- NeurIPS 2026VL-DocIR: A Benchmark for Vision-Based Long Document Retrieval
- NeurIPS 2026Volatility-Whitened Probabilistic Residual Modeling for Long-Term Time Series Forecasting
- NeurIPS 2026What Remains in Sight? Autoregressive Video Decoding as Representation-Guided Context Rewriting
- NeurIPS 2026Why Do Time Series Models Need Long Context Windows?
- NeurIPS 2026Why Transformers Struggle with Distribution-Independent In-Context Learning
在 Claude、Cursor 里直接问这个趋势
接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。
可以直接复制这句话问 AI
用 paperlist 查长上下文在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？
AI 会调用的工具：get_topic_stats {"topic": "long-context"}
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。