上下文压缩论文趋势 2023–2026

主要结论

• 上下文压缩论文从 2023 年的 39 篇增长到 2026 年的 136 篇，三年间扩大到约 3.5 倍。
• NeurIPS 是发上下文压缩论文最多的会议，2023–2026 年共 123 篇，占四个会议总量的 34%。
• 交叉方向里涨得最快的是深度学习 · large language models：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 13 篇。
• 在「长上下文与记忆」的 5 个主题里，它的 2026 年论文数排第 4；同组增长最快的检索记忆是 6.6 倍，上下文压缩是 3.5 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 2026A Free Lunch in LLM Compression: Revisiting Retraining after Pruning
• ICML 2026A Unified Sparse Attention via Multi-Granularity Compression
• ICLR 2026An evolutionary perspective on modes of learning in Transformers
• CVPR 2026AstraNav-Memory Contexts Compression for Long Memory
• NeurIPS 2025AdmTree: Compressing Lengthy Context with Adaptive Semantic Trees
• ICML 2025Does learning the right latent variables necessarily improve in-context learning?
• ICLR 2025Approaching Rate-Distortion Limits in Neural Compression with Lattice Transform Coding
• CVPR 2025ECVC: Exploiting Non-Local Correlations in Multiple Frames for Contextual Video Compression
• NeurIPS 2024StreamingDialogue: Prolonged Dialogue Learning via Long Context Compression with Minimal Losses
• ICML 2024Agnostic Sample Compression Schemes for Regression
• ICLR 2024Compressed Context Memory for Online Language Model Interaction
• CVPR 2024Diversity-aware Channel Pruning for StyleGAN Compression
• NeurIPS 2023Do Not Marginalize Mechanisms, Rather Consolidate!
• ICML 2023Private Federated Learning with Autotuned Compression
• ICLR 2023$k$NN Prompting: Beyond-Context Learning with Calibration-Free Nearest Neighbor Inference
• CVPR 2023AutoAD: Movie Description in Context
• NeurIPS 2026A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
• ICML 2026Context-Driven Incremental Compression for Multi-Turn Dialogue Generation
• ICLR 2026Autoencoding-Free Context Compression for LLMs via Contextual Semantic Anchors
• CVPR 2026Featurising Pixels from Dynamic 3D Scenes with Linear In-Context Learners

全部上下文压缩论文

共 50 篇，第 1 / 1 页

1. NeurIPS 2026A Free Lunch in LLM Compression: Revisiting Retraining after Pruning
2. NeurIPS 2026A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
3. NeurIPS 2026Accelerating Long-Context LLM Prefill via Layer-wise Progressive Token Pruning in Local Deployment
4. NeurIPS 2026Adaptive Mass-Segmented KV Compression for Long-Context Reasoning
5. NeurIPS 2026Attention as In-Context Empirical Bayes: A Two-Stage View via Particle Dynamics
6. NeurIPS 2026Bayes-Sufficient Compression Is Not Enough: How Communication Helps in Multi-Agent Systems?
7. NeurIPS 2026Can In-Context Learning Support Intrinsic Curiosity?
8. NeurIPS 2026Can Language Models Actually Retrieve In-Context? Drowning in Documents at Million Token Scale
9. NeurIPS 2026Clapping: Removing Per-sample Storage for Pipeline Parallel Learning with Communication Compression
10. NeurIPS 2026COMPOT: Calibration-Optimized Matrix Procrustes Orthogonalization for Transformers Compression
11. NeurIPS 2026CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching
12. NeurIPS 2026CutAttn: Discovering Cognitive Transition Layers for Efficient Long-Context Prefilling
13. NeurIPS 2026Does Compression Imply Generalization? A Minimum Description Length Perspective
14. NeurIPS 2026Effective Context in Transformers: An Analysis of Fragmentation and Tokenization
15. NeurIPS 2026FFTSparse: Relative-Position Correlation Guided Sparse Attention for Long-Context Language Models
16. NeurIPS 2026Fix the Structural Bottleneck: Context Compression via Explicit Information Transmission
17. NeurIPS 2026FracEncoder: Towards Adaptive Cognitive Trajectories via Fractional-Order Context Encoding
18. NeurIPS 2026Gaussian Splatting-based Volumetric Video Compression with Sparse 4D Anchors
19. NeurIPS 2026Generalization Without Compression Penalty: A Stability Analysis of Error Feedback
20. NeurIPS 2026GeneZip: Region-Aware Compression for Long Context DNA Modeling
21. NeurIPS 2026Geometry Matters in Packed 3D Attention
22. NeurIPS 2026Git Context Controller: Manage the Context of Agents by Agentic Git
23. NeurIPS 2026High-arity Sample Compression
24. NeurIPS 2026In-context learning to predict critical transitions in dynamical systems
25. NeurIPS 2026Inadvertent Context Leakage in Language Models
26. NeurIPS 2026Incremental Learning in Transformers for In-Context Associative Recall
27. NeurIPS 2026Large Language Models Develop Belief State Geometry In-Context
28. NeurIPS 2026Leech Lattice Vector Quantization for Efficient LLM Compression
29. NeurIPS 2026LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
30. NeurIPS 2026Logarithmic Depth Suffices for In-Context Gradient Descent
31. NeurIPS 2026Logistic Bandits with $\tilde{O}(\sqrt{dT})$ Regret without Context Diversity Assumptions
32. NeurIPS 2026Make Each Token Count: Towards Improving Long-Context Performance with KV Cache Eviction
33. NeurIPS 2026MoEZip: Routing-Aware KV Cache Compression for Sparse Mixture-of-Experts LLMs
34. NeurIPS 2026NyoomFloat12: Accelerating LLM Inference via Lossless 12-bit Weight Compression
35. NeurIPS 2026On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models
36. NeurIPS 2026Proportionality in Ranking Compression
37. NeurIPS 2026RankVQ: Low-Rank Parameterized Commutative Vector Quantization for KV Cache Compression
38. NeurIPS 2026Residual-Autoregressive Context for 3D Gaussian Splatting Compression
39. NeurIPS 2026ResKV: Residual-based Channel-wise Unstructured Pruning for KV Cache Compression
40. NeurIPS 2026Short-Context Dominance: How Much Local Context Natural Language Actually Needs?
41. NeurIPS 2026SpectralKV: Redundancy-Aware KV Cache Compression via Spectral Coreset Selection
42. NeurIPS 2026SplitZip: Ultra Fast Lossless KV Compression for Disaggregated LLM Serving
43. NeurIPS 2026STAC-R: Subspace-Tracked Activation Compression with Residuals
44. NeurIPS 2026T-ReX: Learning Tile-Reuse Indexes for Structured Model Compression
45. NeurIPS 2026The Price of Choosing Examples: Context Overfitting in Adaptive In-Context Learning
46. NeurIPS 2026Transformers Can Learn Multiclass Classification In-Context: Isotropy Governs Generalization
47. NeurIPS 2026Visual Text Compression as Measure Transport
48. NeurIPS 2026When Less is More: The LLM Scaling Paradox in Context Compression
49. NeurIPS 2026Why Do Time Series Models Need Long Context Windows?
50. NeurIPS 2026Why Transformers Struggle with Distribution-Independent In-Context Learning

常见问题

2023–2026 年四大会议一共有多少篇上下文压缩论文？

共 365 篇，按年份分别是 2023 年 39 篇、2024 年 81 篇、2025 年 109 篇、2026 年 136 篇。

NeurIPS 2026 有多少篇上下文压缩论文？

NeurIPS 2026 有 50 篇上下文压缩论文，上一届是 38 篇。

ICLR 2026 有多少篇上下文压缩论文？

ICLR 2026 有 28 篇上下文压缩论文，上一届是 32 篇。

ICML 2026 有多少篇上下文压缩论文？

ICML 2026 有 39 篇上下文压缩论文，上一届是 20 篇。

CVPR 2026 有多少篇上下文压缩论文？

CVPR 2026 有 19 篇上下文压缩论文，上一届是 19 篇。

哪个会议的上下文压缩论文最多？

NeurIPS，2023–2026 年共 123 篇。