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

每年论文数

各会议论文数

按年份和会议统计

长上下文论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
20231818151566
202452436021176
202598875317255
2026130829831341

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

全部长上下文论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 341 篇，第 4 / 7 页

  1. ICML 2026LIFT: A Novel Framework for Enhancing Long-Context Understanding of LLMs via Long Input Fine-Tuning
  2. ICML 2026LIVE: Long-horizon Interactive Video World Modeling
  3. ICML 2026LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth
  4. ICML 2026Long-Context Modeling with Dynamic Hierarchical Sparse Attention for Memory-Constrained LLM Inference
  5. ICML 2026Long-Horizon Model-Based Offline Reinforcement Learning Without Explicit Conservatism
  6. ICML 2026LoSA: Locality Aware Sparse Attention in Diffusion Language Models
  7. ICML 2026Lost in Context: Adressing Context Anxiety in Large Language Models
  8. ICML 2026Many-Shot CoT-ICL: Making In-Context Learning Truly Learn
  9. ICML 2026Masks Can Be Distracting: On Context Comprehension in Diffusion Language Models
  10. ICML 2026MEG-XL: Data-Efficient Brain-to-Text via Long-Context Pre-Training
  11. ICML 2026Meta Context Engineering via Agentic Skill Evolution
  12. ICML 2026MGAL: A Multilingual Granularity-Aware Long-Context Benchmark
  13. ICML 2026Modeling Long-Tail Relations in the Operating Room via In-Context Multimodal Learning
  14. ICML 2026MoRGen: Mixture-of-Resolutions Generative Forecasting for Irregularly Sampled Medical Time-Series Data
  15. ICML 2026MuonSSM: Orthogonalizing State Space Models for Sequence Modeling
  16. ICML 2026NanoSpec: Accelerating Speculative Decoding using Minimalist In-Context Vocabularies
  17. ICML 2026OBCache: Optimal Brain KV Cache Pruning for Efficient Long-Context LLM Inference
  18. ICML 2026OPIC: Enhancing Language Model Merging via Optimizing In-Context Capability
  19. ICML 2026ParaTool: Shifting Tool Representations from Context to Parameters
  20. ICML 2026ParisKV: Fast and Drift-Robust KV-Cache Retrieval for Long-Context LLMs
  21. ICML 2026PersistBench: When Should Long-Term Memories Be Forgotten by LLMs?
  22. ICML 2026PLASH: Provably Linear-Time Attention with Selective Higher-Order Feature Sketching
  23. ICML 2026Predicting Future Utility: Global Combinatorial Optimization for Task-Agnostic KV Cache Eviction
  24. ICML 2026Prism: Spectral-Aware Block-Sparse Attention
  25. ICML 2026Probing How Scalable Table Data Enhances General Long-Context Reasoning
  26. ICML 2026ProphetKV: User-Query-Driven Selective Recomputation for Efficient KV Cache Reuse in Retrieval-Augmented Generation
  27. ICML 2026Push, Pop, Parallelize: Stack-Augmented Linear Attention via the Delta Rule
  28. ICML 2026RaBitQCache: Rotated Binary Quantization for KVCache in Long Context LLM Inference
  29. ICML 2026Recursive Models for Long-Horizon Reasoning
  30. ICML 2026RePo: Language Models with Context Re-Positioning
  31. ICML 2026Residual Context Diffusion Language Models
  32. ICML 2026Resolving the Timestep Scaling Paradox in Spiking Neural Networks with a Timestep-Scalable Neuron Model
  33. ICML 2026Rethinking Genomic Modeling Through Optical Character Recognition
  34. ICML 2026Rethinking Loss Reweighting for Imbalance Learning as an Inverse Problem: A Neural Collapse Point of View
  35. ICML 2026Revealing Long-context Potential of Attention Heads via Frequency Kernels
  36. ICML 2026Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation
  37. ICML 2026Safe In-Context Reinforcement Learning
  38. ICML 2026SALE : Low-bit Estimation for Efficient Sparse Attention in Long-context LLM Prefilling
  39. ICML 2026Scalable Option Learning in High-Throughput Environments
  40. ICML 2026Scaling Long-Horizon Agent via Context Folding
  41. ICML 2026Scout Before You Attend: Sketch-and-Walk Sparse Attention for Efficient LLM Inference
  42. ICML 2026Shortcut-Resistant CAM Distillation for Long-Tailed Recognition
  43. ICML 2026Shuffle the Context: RoPE-Perturbed Self-Distillation for Long-Context Adaptation
  44. ICML 2026SKETCH: Semantic Key-Point Conditioning for Long-Horizon Vessel Trajectory Prediction
  45. ICML 2026Softmax as Linear Attention in the Large-Prompt Regime: a Measure-based Perspective
  46. ICML 2026Softplus Attention with Re-weighting Boosts Length Extrapolation in Large Language Models
  47. ICML 2026SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?
  48. ICML 2026Symmetry Reveals Layerwise Dynamics: How Transformers Perform In-Context Classification
  49. ICML 2026Temporal Context Reinstatement Drives Episodic-Like Order Memory in Long-Context Language Models
  50. ICML 2026The Devil is in the Spectrum: Mitigating Representation Collapse in LLMs via Topologically Regularized Side-Path

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"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

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。

「长上下文与记忆」下的其他主题