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 298 篇，第 5 / 6 页

  1. NeurIPS 2025Scale-invariant attention
  2. NeurIPS 2025SeerAttention: Self-distilled Attention Gating for Efficient Long-context Prefilling
  3. NeurIPS 2025Short-length Adversarial Training Helps LLMs Defend Long-length Jailbreak Attacks: Theoretical and Empirical Evidence
  4. NeurIPS 2025Skrull: Towards Efficient Long Context Fine-tuning through Dynamic Data Scheduling
  5. NeurIPS 2025SkyLadder: Better and Faster Pretraining via Context Window Scheduling
  6. NeurIPS 2025StarTrail: Concentric Ring Sequence Parallelism for Efficient Near-Infinite-Context Transformer Model Training
  7. NeurIPS 2025StateSpaceDiffuser: Bringing Long Context to Diffusion World Models
  8. NeurIPS 2025Streaming Attention Approximation via Discrepancy Theory
  9. NeurIPS 2025Technical Debt in In-Context Learning: Diminishing Efficiency in Long Context
  10. NeurIPS 2025Temporal Chain of Thought: Long-Video Understanding by Thinking in Frames
  11. NeurIPS 2025Theoretical Insights into In-context Learning with Unlabeled Data
  12. NeurIPS 2025Thoughts Are All Over the Place: On the Underthinking of Long Reasoning Models
  13. NeurIPS 2025TiRex: Zero-Shot Forecasting Across Long and Short Horizons with Enhanced In-Context Learning
  14. NeurIPS 2025Titans: Learning to Memorize at Test Time
  15. NeurIPS 2025Towards Predicting Any Human Trajectory In Context
  16. NeurIPS 2025Towards Provable Emergence of In-Context Reinforcement Learning
  17. NeurIPS 2025Trained Mamba Emulates Online Gradient Descent in In-Context Linear Regression
  18. NeurIPS 2025True Zero-Shot Inference of Dynamical Systems Preserving Long-Term Statistics
  19. NeurIPS 2025UniGist: Towards General and Hardware-aligned Sequence-level Long Context Compression
  20. NeurIPS 2025Unlabeled Data Can Provably Enhance In-Context Learning of Transformers
  21. NeurIPS 2025Unleashing Hour-Scale Video Training for Long Video-Language Understanding
  22. NeurIPS 2025Unveiling the Spatial-temporal Effective Receptive Fields of Spiking Neural Networks
  23. NeurIPS 2025Variance-Reduced Long-Term Rehearsal Learning with Quadratic Programming Reformulation
  24. NeurIPS 2025Vision-centric Token Compression in Large Language Model
  25. NeurIPS 2025Vocabulary In-Context Learning in Transformers: Benefits of Positional Encoding
  26. NeurIPS 2025WaLRUS: Wavelets for Long range Representation Using State Space Methods
  27. NeurIPS 2025WorldMem: Long-term Consistent World Simulation with Memory
  28. NeurIPS 2025ZeCO: Zero-Communication Overhead Sequence Parallelism for Linear Attention
  29. NeurIPS 2024[Re] CUDA: Curriculum of Data Augmentation for Long‐tailed Recognition
  30. NeurIPS 2024An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encoding
  31. NeurIPS 2024AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation
  32. NeurIPS 2024BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack
  33. NeurIPS 2024Balancing Context Length and Mixing Times for Reinforcement Learning at Scale
  34. NeurIPS 2024Base of RoPE Bounds Context Length
  35. NeurIPS 2024BERTs are Generative In-Context Learners
  36. NeurIPS 2024DAPE: Data-Adaptive Positional Encoding for Length Extrapolation
  37. NeurIPS 2024Dataset Decomposition: Faster LLM Training with Variable Sequence Length Curriculum
  38. NeurIPS 2024Decision Mamba: Reinforcement Learning via Hybrid Selective Sequence Modeling
  39. NeurIPS 2024Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data
  40. NeurIPS 2024Efficient Recurrent Off-Policy RL Requires a Context-Encoder-Specific Learning Rate
  41. NeurIPS 2024Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical Perspective
  42. NeurIPS 2024Exploring Context Window of Large Language Models via Decomposed Positional Vectors
  43. NeurIPS 2024Flexible Context-Driven Sensory Processing in Dynamical Vision Models
  44. NeurIPS 2024Found in the Middle: How Language Models Use Long Contexts Better via Plug-and-Play Positional Encoding
  45. NeurIPS 2024Grammar-Aligned Decoding
  46. NeurIPS 2024In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness
  47. NeurIPS 2024InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory
  48. NeurIPS 2024KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
  49. NeurIPS 2024Learning Partitions from Context
  50. NeurIPS 2024Lever LM: Configuring In-Context Sequence to Lever Large Vision Language Models

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