数据集与基准论文趋势 2023–2026

新数据集与评测协议

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 5,856 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

数据集与基准论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
2023227162166382937
20244482163183641,346
20256374191893371,582
20266236352754581,991

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

全部数据集与基准论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 948 篇，第 9 / 19 页

  1. ICML 2025Position: In-House Evaluation Is Not Enough. Towards Robust Third-Party Evaluation and Flaw Disclosure for General-Purpose AI
  2. ICML 2025Position: Medical Large Language Model Benchmarks Should Prioritize Construct Validity
  3. ICML 2025Position: Societal Impacts Research Requires Benchmarks for Creative Composition Tasks
  4. ICML 2025Position: Stop treating `AGI' as the north-star goal of AI research
  5. ICML 2025Position: The Categorization of Race in ML is a Flawed Premise
  6. ICML 2025Position: Theory of Mind Benchmarks are Broken for Large Language Models
  7. ICML 2025Position: Truly Self-Improving Agents Require Intrinsic Metacognitive Learning
  8. ICML 2025Positive-unlabeled AUC Maximization under Covariate Shift
  9. ICML 2025Potemkin Understanding in Large Language Models
  10. ICML 2025Predicting the Susceptibility of Examples to Catastrophic Forgetting
  11. ICML 2025Predictive Data Selection: The Data That Predicts Is the Data That Teaches
  12. ICML 2025Prompt-based Depth Pruning of Large Language Models
  13. ICML 2025Protein Structure Tokenization: Benchmarking and New Recipe
  14. ICML 2025QEM-Bench: Benchmarking Learning-based Quantum Error Mitigation and QEMFormer as a Multi-ranged Context Learning Baseline
  15. ICML 2025RAGGED: Towards Informed Design of Scalable and Stable RAG Systems
  16. ICML 2025Reconstructing Cell Lineage Trees from Phenotypic Features with Metric Learning
  17. ICML 2025Regression for the Mean: Auto-Evaluation and Inference with Few Labels through Post-hoc Regression
  18. ICML 2025Reliable and Efficient Amortized Model-based Evaluation
  19. ICML 2025Rényi Neural Processes
  20. ICML 2025Rethinking the Stability-Plasticity Trade-off in Continual Learning from an Architectural Perspective
  21. ICML 2025Robust ML Auditing using Prior Knowledge
  22. ICML 2025Robust Offline Reinforcement Learning with Linearly Structured $f$-Divergence Regularization
  23. ICML 2025ROME is Forged in Adversity: Robust Distilled Datasets via Information Bottleneck
  24. ICML 2025Safe Delta: Consistently Preserving Safety when Fine-Tuning LLMs on Diverse Datasets
  25. ICML 2025Sanity Checking Causal Representation Learning on a Simple Real-World System
  26. ICML 2025Scalable Approximation Algorithms for $p$-Wasserstein Distance and Its Variants
  27. ICML 2025SEFE: Superficial and Essential Forgetting Eliminator for Multimodal Continual Instruction Tuning
  28. ICML 2025Semantic Shift Estimation via Dual-Projection and Classifier Reconstruction for Exemplar-Free Class-Incremental Learning
  29. ICML 2025SIMPLEMIX: Frustratingly Simple Mixing of Off- and On-policy Data in Language Model Preference Learning
  30. ICML 2025SLiM: One-shot Quantization and Sparsity with Low-rank Approximation for LLM Weight Compression
  31. ICML 2025SNS-Bench: Defining, Building, and Assessing Capabilities of Large Language Models in Social Networking Services
  32. ICML 2025Solving Probabilistic Verification Problems of Neural Networks using Branch and Bound
  33. ICML 2025Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization
  34. ICML 2025Stable Offline Value Function Learning with Bisimulation-based Representations
  35. ICML 2025STAMP Your Content: Proving Dataset Membership via Watermarked Rephrasings
  36. ICML 2025Stochastic Forward–Backward Deconvolution: Training Diffusion Models with Finite Noisy Datasets
  37. ICML 2025TabPFN Unleashed: A Scalable and Effective Solution to Tabular Classification Problems
  38. ICML 2025TabSDS: a Lightweight, Fully Non-Parametric, and Model Free Approach for Generating Synthetic Tabular Data
  39. ICML 2025Taming Diffusion for Dataset Distillation with High Representativeness
  40. ICML 2025TeDS: Joint Learning of Diachronic and Synchronic Perspectives in Quaternion Space for Temporal Knowledge Graph Completion
  41. ICML 2025The Best of Both Worlds: Bridging Quality and Diversity in Data Selection with Bipartite Graph
  42. ICML 2025The Emperor's New Clothes in Benchmarking? A Rigorous Examination of Mitigation Strategies for LLM Benchmark Data Contamination
  43. ICML 2025The Power of Random Features and the Limits of Distribution-Free Gradient Descent
  44. ICML 2025The Price of Freedom: Exploring Expressivity and Runtime Tradeoffs in Equivariant Tensor Products
  45. ICML 2025The Underlying Universal Statistical Structure of Natural Datasets
  46. ICML 2025Three-Dimensional Trajectory Prediction with 3DMoTraj Dataset
  47. ICML 2025Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage
  48. ICML 2025Towards a General Time Series Forecasting Model with Unified Representation and Adaptive Transfer
  49. ICML 2025Training Software Engineering Agents and Verifiers with SWE-Gym
  50. ICML 2025Trajectory World Models for Heterogeneous Environments

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 查数据集与基准在 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": "dataset-and-benchmarks"}

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数据集与基准论文？
共 5,856 篇，按年份分别是 2023 年 937 篇、2024 年 1,346 篇、2025 年 1,582 篇、2026 年 1,991 篇。
NeurIPS 2026 有多少篇数据集与基准论文？
NeurIPS 2026 有 623 篇数据集与基准论文，上一届是 637 篇。
ICLR 2026 有多少篇数据集与基准论文？
ICLR 2026 有 635 篇数据集与基准论文，上一届是 419 篇。
ICML 2026 有多少篇数据集与基准论文？
ICML 2026 有 275 篇数据集与基准论文，上一届是 189 篇。
CVPR 2026 有多少篇数据集与基准论文？
CVPR 2026 有 458 篇数据集与基准论文，上一届是 337 篇。
哪个会议的数据集与基准论文最多？
NeurIPS，2023–2026 年共 1,935 篇。

「效率与系统」下的其他主题