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 838 篇，第 17 / 17 页

801. ICLR 2023On the Effectiveness of Out-of-Distribution Data in Self-Supervised Long-Tail Learning.
802. ICLR 2023Simplified State Space Layers for Sequence Modeling
803. ICLR 2023Tier Balancing: Towards Dynamic Fairness over Underlying Causal Factors
804. ICLR 2023What learning algorithm is in-context learning? Investigations with linear models
805. ICLR 2023What Makes Convolutional Models Great on Long Sequence Modeling?
806. ICML 2023A Critical View of Vision-Based Long-Term Dynamics Prediction Under Environment Misalignment
807. ICML 2023CAB: Comprehensive Attention Benchmarking on Long Sequence Modeling
808. ICML 2023Context Consistency Regularization for Label Sparsity in Time Series
809. ICML 2023Controlled Differential Equations on Long Sequences via Non-standard Wavelets
810. ICML 2023DevFormer: A Symmetric Transformer for Context-Aware Device Placement
811. ICML 2023FusionRetro: Molecule Representation Fusion via In-Context Learning for Retrosynthetic Planning
812. ICML 2023Global Context Vision Transformers
813. ICML 2023Large Language Models Struggle to Learn Long-Tail Knowledge
814. ICML 2023Long Horizon Temperature Scaling
815. ICML 2023Long-Term Rhythmic Video Soundtracker
816. ICML 2023LongCoder: A Long-Range Pre-trained Language Model for Code Completion
817. ICML 2023Resurrecting Recurrent Neural Networks for Long Sequences
818. ICML 2023Robust Situational Reinforcement Learning in Face of Context Disturbances
819. ICML 2023Simple Hardware-Efficient Long Convolutions for Sequence Modeling
820. ICML 2023Transformers Learn In-Context by Gradient Descent
821. NeurIPS 2023A Single 2D Pose with Context is Worth Hundreds for 3D Human Pose Estimation
822. NeurIPS 2023Blockwise Parallel Transformers for Large Context Models
823. NeurIPS 2023Construction of Hierarchical Neural Architecture Search Spaces based on Context-free Grammars
824. NeurIPS 2023Context-lumpable stochastic bandits
825. NeurIPS 2023Context-PIPs: Persistent Independent Particles Demands Spatial Context Features
826. NeurIPS 2023Digital Typhoon: Long-term Satellite Image Dataset for the Spatio-Temporal Modeling of Tropical Cyclones
827. NeurIPS 2023Explore In-Context Learning for 3D Point Cloud Understanding
828. NeurIPS 2023Focused Transformer: Contrastive Training for Context Scaling
829. NeurIPS 2023HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution
830. NeurIPS 2023Meta-in-context learning in large language models
831. NeurIPS 2023On the Role of Noise in the Sample Complexity of Learning Recurrent Neural Networks: Exponential Gaps for Long Sequences
832. NeurIPS 2023PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solvers
833. NeurIPS 2023PRODIGY: Enabling In-context Learning Over Graphs
834. NeurIPS 2023Recursion in Recursion: Two-Level Nested Recursion for Length Generalization with Scalability
835. NeurIPS 2023SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecasting
836. NeurIPS 2023Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection
837. NeurIPS 2023Unlimiformer: Long-Range Transformers with Unlimited Length Input
838. NeurIPS 2023Would I have gotten that reward? Long-term credit assignment by counterfactual contribution analysis

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。