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 176 篇，第 3 / 4 页

  1. ICML 2024Long Range Propagation on Continuous-Time Dynamic Graphs
  2. ICML 2024Long-Tail Learning with Foundation Model: Heavy Fine-Tuning Hurts
  3. ICML 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
  4. ICML 2024Loss Shaping Constraints for Long-Term Time Series Forecasting
  5. ICML 2024Memory Consolidation Enables Long-Context Video Understanding
  6. ICML 2024Meta-Reinforcement Learning Robust to Distributional Shift Via Performing Lifelong In-Context Learning
  7. ICML 2024PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learning
  8. ICML 2024PolySketchFormer: Fast Transformers via Sketching Polynomial Kernels
  9. ICML 2024Position: Cracking the Code of Cascading Disparity Towards Marginalized Communities
  10. ICML 2024Position: Do pretrained Transformers Learn In-Context by Gradient Descent?
  11. ICML 2024Position: Key Claims in LLM Research Have a Long Tail of Footnotes
  12. ICML 2024QUEST: Query-Aware Sparsity for Efficient Long-Context LLM Inference
  13. ICML 2024Rethinking Adversarial Robustness in the Context of the Right to be Forgotten
  14. ICML 2024Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences
  15. ICML 2024SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning
  16. ICML 2024SIN: Selective and Interpretable Normalization for Long-Term Time Series Forecasting
  17. ICML 2024SparseTSF: Modeling Long-term Time Series Forecasting with *1k* Parameters
  18. ICML 2024Training-Free Long-Context Scaling of Large Language Models
  19. ICML 2024Transformers Implement Functional Gradient Descent to Learn Non-Linear Functions In Context
  20. ICML 2024Transformers Learn Nonlinear Features In Context: Nonconvex Mean-field Dynamics on the Attention Landscape
  21. ICML 2024Viewing Transformers Through the Lens of Long Convolutions Layers
  22. ICML 2024What is the Long-Run Distribution of Stochastic Gradient Descent? A Large Deviations Analysis
  23. ICML 2024Why Larger Language Models Do In-context Learning Differently?
  24. ICML 2024xT: Nested Tokenization for Larger Context in Large Images
  25. NeurIPS 2024[Re] CUDA: Curriculum of Data Augmentation for Long‐tailed Recognition
  26. NeurIPS 2024An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encoding
  27. NeurIPS 2024AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation
  28. NeurIPS 2024BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack
  29. NeurIPS 2024Balancing Context Length and Mixing Times for Reinforcement Learning at Scale
  30. NeurIPS 2024Base of RoPE Bounds Context Length
  31. NeurIPS 2024BERTs are Generative In-Context Learners
  32. NeurIPS 2024DAPE: Data-Adaptive Positional Encoding for Length Extrapolation
  33. NeurIPS 2024Dataset Decomposition: Faster LLM Training with Variable Sequence Length Curriculum
  34. NeurIPS 2024Decision Mamba: Reinforcement Learning via Hybrid Selective Sequence Modeling
  35. NeurIPS 2024Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data
  36. NeurIPS 2024Efficient Recurrent Off-Policy RL Requires a Context-Encoder-Specific Learning Rate
  37. NeurIPS 2024Enhancing In-Context Learning Performance with just SVD-Based Weight Pruning: A Theoretical Perspective
  38. NeurIPS 2024Exploring Context Window of Large Language Models via Decomposed Positional Vectors
  39. NeurIPS 2024Flexible Context-Driven Sensory Processing in Dynamical Vision Models
  40. NeurIPS 2024Found in the Middle: How Language Models Use Long Contexts Better via Plug-and-Play Positional Encoding
  41. NeurIPS 2024Grammar-Aligned Decoding
  42. NeurIPS 2024In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness
  43. NeurIPS 2024InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory
  44. NeurIPS 2024KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
  45. NeurIPS 2024Learning Partitions from Context
  46. NeurIPS 2024Lever LM: Configuring In-Context Sequence to Lever Large Vision Language Models
  47. NeurIPS 2024Linear Transformers are Versatile In-Context Learners
  48. NeurIPS 2024Long-range Brain Graph Transformer
  49. NeurIPS 2024Long-Range Feedback Spiking Network Captures Dynamic and Static Representations of the Visual Cortex under Movie Stimuli
  50. NeurIPS 2024Long-tailed Object Detection Pretraining: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction

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