Long Context papers, 2023–2026

Key findings

• Long Context went from 66 papers in 2023 to 341 in 2026, roughly 5.2x in three years.
• NeurIPS publishes the most Long Context work: 298 papers from 2023 to 2026, 36% of the four-venue total.
• The fastest-growing overlap is with AI Agents: papers tagged with both rose from 0 in 2023 to 45 in 2026.
• Among the 5 topics under Long Context and Memory, it ranks #1 by 2026 paper count; the fastest grower in the group, Retrieval Memory, is up 6.6x versus 5.2x here.

Representative papers

Spread across years and venues so every edition shows up

• 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

All Long Context papers

838 papers, page 5 of 17

201. ICML 2026The First Drop of Ink: Nonlinear Impact of Misleading Information in Long-Context Reasoning
202. ICML 2026The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning
203. ICML 2026Token Sparse Attention: Efficient Long-Context Inference with Interleaved Token Selection
204. ICML 2026Training–Inference Consistent Segmented Execution for Long-Context LLMs
205. ICML 2026Transformers Can Learn Posterior Predictive Distributions In-Context
206. ICML 2026Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent
207. ICML 2026Understanding Generalization and Forgetting in In-Context Continual Learning
208. ICML 2026Universal Approximation with Softmax Attention
209. ICML 2026Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking
210. ICML 2026When to Memorize and When to Stop: Gated Recurrent Memory for Long-Context Reasoning
211. ICML 2026You Need Better Attention Priors
212. NeurIPS 2026$\boldsymbol{f}$-OPD: Stabilizing Long-Horizon On-Policy Distillation with Freshness-Aware Control
213. NeurIPS 2026A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
214. NeurIPS 2026A Solvable Model of Chain-of-Thought in In-Context Learning
215. NeurIPS 2026Accelerating Long-Context LLM Prefill via Layer-wise Progressive Token Pruning in Local Deployment
216. NeurIPS 2026AdaCal: Adaptive Calibration for Robust Sparse Attention in Long-Context LLMs
217. NeurIPS 2026Adaptive Mass-Segmented KV Compression for Long-Context Reasoning
218. NeurIPS 2026Align-RAG: Alignment Is All You Need for TSFM In-Context Learning
219. NeurIPS 2026Ask Early, Ask Late, Ask Right: When Does Clarification Timing Matter for Long-Horizon Agents?
220. NeurIPS 2026Attention as In-Context Empirical Bayes: A Two-Stage View via Particle Dynamics
221. NeurIPS 2026Attention Drift: What Auto-Regressive Speculative Decoding Models Learn
222. NeurIPS 2026Attention Itself Could Retrieve. RetrieveVGGT: Training-Free Long Context Streaming 3D Reconstruction via Query-Key Similarity Retrieval
223. NeurIPS 2026Beyond Suspicious Steps: Ontological Trust in Long-Horizon Agents
224. NeurIPS 2026Block Sparse Flash Attention
225. NeurIPS 2026Block-Wise Differentiable Sinkhorn Attention: Tail-Refinement Gradients with a Gap-Aware Dustbin Bridge
226. NeurIPS 2026Breadcrumbing Search Agents: Per-Turn Scheming Over Long-Horizon Trajectories
227. NeurIPS 2026Bridging Risk Approximation Gaps in Model Predictive Task Sampling via In-Context Modeling
228. NeurIPS 2026Can In-Context Learning Support Intrinsic Curiosity?
229. NeurIPS 2026Can Language Models Actually Retrieve In-Context? Drowning in Documents at Million Token Scale
230. NeurIPS 2026CEO-Bench: Can Agents Play the Long Game?
231. NeurIPS 2026Choosing Before Acting: Comparative Value Estimation for Long-Horizon Tool-Use Agents
232. NeurIPS 2026CivBench: A Long-Horizon Benchmark for Tool-Mediated Agents in Civilization VI
233. NeurIPS 2026CLoSeR: Closing the Loop for Long-Context Streaming Reconstruction
234. NeurIPS 2026CoLaX: Context-Aware Local Explanations for Time Series Classification
235. NeurIPS 2026CommunityKV: Efficient Long-Context Decoding via Graph Partitioning
236. NeurIPS 2026CompactAttention: Accelerating Chunked Prefill with Block-Union KV Selection
237. NeurIPS 2026Constrained Modulatory Reservoirs for Context-Dependent Computation
238. NeurIPS 2026Context as Low-Rank Weights: Bounded Parametric Dynamic Memory for Unbounded Context
239. NeurIPS 2026ContextShift: A Controlled Benchmark for Context Dependence in Object Detection
240. NeurIPS 2026Controlling Transient Amplification Improves Long-horizon Rollouts
241. NeurIPS 2026Correspondence Pruning by Iterative Structural Rectification
242. NeurIPS 2026CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching
243. NeurIPS 2026CutAttn: Discovering Cognitive Transition Layers for Efficient Long-Context Prefilling
244. NeurIPS 2026Do Enterprise Systems Need Learned World Models? The Importance of Context to Infer Dynamics
245. NeurIPS 2026DSR-TSF: Spectrum-Driven Dynamic Routing for Efficient Long-Horizon Time Series Forecasting
246. NeurIPS 2026DualKV: Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts
247. NeurIPS 2026Dynamic Context Modeling for Longitudinal Mental Health Monitoring under Distribution Shift
248. NeurIPS 2026EAT: Eviction-Aware Training for Long-Context LLM Inference on Edge Devices
249. NeurIPS 2026Effective Context in Transformers: An Analysis of Fragmentation and Tokenization
250. NeurIPS 2026Efficient Scaling of LLM Training with Flexible Context Parallelism

FAQ

How many Long Context papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

838 in total: 66 in 2023, 176 in 2024, 255 in 2025, 341 in 2026.

How many Long Context papers are at NeurIPS 2026?

NeurIPS 2026 has 130 Long Context papers, versus 98 the year before.

How many Long Context papers are at ICLR 2026?

ICLR 2026 has 82 Long Context papers, versus 87 the year before.

How many Long Context papers are at ICML 2026?

ICML 2026 has 98 Long Context papers, versus 53 the year before.

How many Long Context papers are at CVPR 2026?

CVPR 2026 has 31 Long Context papers, versus 17 the year before.

Which conference publishes the most Long Context papers?

NeurIPS, with 298 papers from 2023 to 2026.