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 84 篇，第 1 / 2 页

1. CVPR 2026AR2-4FV Anchored Referring and Re-identification for Long-Term Grounding in Fixed-View
2. CVPR 2026AstraNav-Memory Contexts Compression for Long Memory
3. CVPR 2026BEV-CAR Enhancing Monocular Birds Eye View Segmentation with Context-Aware Rasterization
4. CVPR 2026CDICS Delving Into Fine-Grained Attribute for In-Context Segmentation via Compositional
5. CVPR 2026CHIRP dataset towards long-term individual-level behavioral monitoring of bird populations
6. CVPR 2026Common Inpainted Objects In-N-Out of Context
7. CVPR 2026Deformation-based In-Context Learning for Point Cloud Understanding
8. CVPR 2026Expanding Spatial and Temporal Context for Robotic Imitation Learning With
9. CVPR 2026Featurising Pixels from Dynamic 3D Scenes with Linear In-Context Learners
10. CVPR 2026FedCART Tackling Long-Tailed Distributions in Federated Adversarial Training via Classifier
11. CVPR 2026Fine-Tuning Impairs the Balancedness of Foundation Models in Long-tailed Personalized
12. CVPR 2026FoSS Modeling Long-Range Dependencies and Multimodal Uncertainty in Trajectory Prediction
13. CVPR 2026INSID3 Training-Free In-Context Segmentation with DINOv3
14. CVPR 2026Long-SCOPE Fully Sparse Long-Range Cooperative 3D Perception
15. CVPR 2026Long-Tail Internet Photo Reconstruction
16. CVPR 2026Meta-Learning In-Context Enables Training-Free Cross Subject Brain Decoding
17. CVPR 2026MoRel Long-Range Flicker-Free 4D Motion Modeling via Anchor Relay-based Bidirectioanl
18. CVPR 2026MuKV Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering
19. CVPR 2026Perceiving the Near Reasoning the Distant Coherent Long-Horizon Trajectory Prediction
20. CVPR 2026PersonaVLM Long-Term Personalized Multimodal LLMs
21. CVPR 2026PR-MaGIC Prompt Refinement Via Mask Decoder Gradient Flow For In-Context
22. CVPR 2026Recurrent Reasoning with Vision-Language Models for Estimating Long-Horizon Embodied Task
23. CVPR 2026Reframing Long-Tailed Learning via Loss Landscape Geometry
24. CVPR 2026Seeing Conversations Communication Context Identification in Egocentric Video
25. CVPR 2026Semantic Context Matters Improving Conditioning for Autoregressive Models
26. CVPR 2026SemLT3D Semantic-Guided Expert Distillation for Camera-only Long-Tailed 3D Object Detection
27. CVPR 2026Taming the Long Tail Rebalancing Adversarial Training via Adaptive Perturbation
28. CVPR 2026tttLRM Test-Time Training for Long Context and Autoregressive 3D Reconstruction
29. CVPR 2026VecAttention Vector-wise Sparse Attention for Accelerating Long Context Inference
30. CVPR 2026Video Panels for Long Video Understanding
31. CVPR 2026Wavelet-based Frame Selection by Detecting Semantic Boundary for Long Video
32. CVPR 2025BIMBA: Selective-Scan Compression for Long-Range Video Question Answering
33. CVPR 2025Chapter-Llama: Efficient Chaptering in Hour-Long Videos with LLMs
34. CVPR 2025Context-Enhanced Memory-Refined Transformer for Online Action Detection
35. CVPR 2025Distilling Long-tailed Datasets
36. CVPR 2025Fractal Calibration for Long-tailed Object Detection
37. CVPR 2025From Head to Tail: Efficient Black-box Model Inversion Attack via Long-tailed Learning
38. CVPR 2025GA3CE: Unconstrained 3D Gaze Estimation with Gaze-Aware 3D Context Encoding
39. CVPR 2025HOIGPT: Learning Long-Sequence Hand-Object Interaction with Language Models
40. CVPR 2025Learning from Neighbors: Category Extrapolation for Long-Tail Learning
41. CVPR 2025LOCORE: Image Re-ranking with Long-Context Sequence Modeling
42. CVPR 2025Lost in Translation, Found in Context: Sign Language Translation with Contextual Cues
43. CVPR 2025Mimic In-Context Learning for Multimodal Tasks
44. CVPR 2025Neural Video Compression with Context Modulation
45. CVPR 2025OverLoCK: An Overview-first-Look-Closely-next ConvNet with Context-Mixing Dynamic Kernels
46. CVPR 2025Precise Event Spotting in Sports Videos: Solving Long-Range Dependency and Class Imbalance
47. CVPR 2025TAET: Two-Stage Adversarial Equalization Training on Long-Tailed Distributions
48. CVPR 2025Tokenize Image Patches: Global Context Fusion for Effective Haze Removal in Large Images
49. CVPR 2024Amodal Completion via Progressive Mixed Context Diffusion
50. CVPR 2024CLiC: Concept Learning in 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。