大模型基准论文趋势 2023–2026

能力评测

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 544 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

大模型基准论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202320002
2024241321361
20259250248174
2026138103606307

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

全部大模型基准论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 544 篇，第 10 / 11 页

  1. NeurIPS 2025Reasoning Is Not a Race: When Stopping Early Beats Going Deeper
  2. NeurIPS 2025Reparameterized LLM Training via Orthogonal Equivalence Transformation
  3. NeurIPS 2025Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
  4. NeurIPS 2025Scale-invariant attention
  5. NeurIPS 2025Self-Generated In-Context Examples Improve LLM Agents for Sequential Decision-Making Tasks
  6. NeurIPS 2025Semantic-KG: Using Knowledge Graphs to Construct Benchmarks for Measuring Semantic Similarity
  7. NeurIPS 2025Set-LLM: A Permutation-Invariant LLM
  8. NeurIPS 2025Shape it Up! Restoring LLM Safety during Finetuning
  9. NeurIPS 2025Sheetpedia: A 300K-Spreadsheet Corpus for Spreadsheet Intelligence and LLM Fine-Tuning
  10. NeurIPS 2025Silencer: From Discovery to Mitigation of Self-Bias in LLM-as-Benchmark-Generator
  11. NeurIPS 2025SmallKV: Small Model Assisted Compensation of KV Cache Compression for Efficient LLM Inference
  12. NeurIPS 2025SmartCache: Context-aware Semantic Cache for Efficient Multi-turn LLM Inference
  13. NeurIPS 2025Spatially-aware Weights Tokenization for NeRF-Language Models
  14. NeurIPS 2025Stackelberg Self-Annotation: A Robust Approach to Data-Efficient LLM Alignment
  15. NeurIPS 2025SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines
  16. NeurIPS 2025Superposition Yields Robust Neural Scaling
  17. NeurIPS 2025SWE-bench Goes Live!
  18. NeurIPS 2025Tail-Optimized Caching for LLM Inference
  19. NeurIPS 2025The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements
  20. NeurIPS 2025The Leaderboard Illusion
  21. NeurIPS 2025The Lighthouse of Language: Enhancing LLM Agents via Critique-Guided Improvement
  22. NeurIPS 2025TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks
  23. NeurIPS 2025Thinkless: LLM Learns When to Think
  24. NeurIPS 2025Toward Engineering AGI: Benchmarking the Engineering Design Capabilities of LLMs
  25. NeurIPS 2025Towards Fully FP8 GEMM LLM Training at Scale
  26. NeurIPS 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning
  27. NeurIPS 2025Trajectory Bellman Residual Minimization: A Simple Value-Based Method for LLM Reasoning
  28. NeurIPS 2025Transformer Copilot: Learning from The Mistake Log in LLM Fine-tuning
  29. NeurIPS 2025WALL-E: World Alignment by NeuroSymbolic Learning improves World Model-based LLM Agents
  30. NeurIPS 2025WritingBench: A Comprehensive Benchmark for Generative Writing
  31. NeurIPS 2025Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibrator
  32. CVPR 2024Honeybee: Locality-enhanced Projector for Multimodal LLM
  33. CVPR 2024Koala: Key Frame-Conditioned Long Video-LLM
  34. CVPR 2024Self-correcting LLM-controlled Diffusion Models
  35. ICLR 2024BEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks
  36. ICLR 2024Can LLM-Generated Misinformation Be Detected?
  37. ICLR 2024ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate
  38. ICLR 2024Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactuals
  39. ICLR 2024LLM-Assisted Code Cleaning For Training Accurate Code Generators
  40. ICLR 2024LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset
  41. ICLR 2024PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization
  42. ICLR 2024PB-LLM: Partially Binarized Large Language Models
  43. ICLR 2024Prometheus: Inducing Fine-Grained Evaluation Capability in Language Models
  44. ICLR 2024SKILL-MIX: a Flexible and Expandable Family of Evaluations for AI Models
  45. ICLR 2024Steve-Eye: Equipping LLM-based Embodied Agents with Visual Perception in Open Worlds
  46. ICLR 2024The LLM Surgeon
  47. ICLR 2024To the Cutoff... and Beyond? A Longitudinal Perspective on LLM Data Contamination
  48. ICML 2024An LLM Compiler for Parallel Function Calling
  49. ICML 2024Any-Precision LLM: Low-Cost Deployment of Multiple, Different-Sized LLMs
  50. ICML 2024Auto-Encoding Morph-Tokens for Multimodal LLM

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 查大模型基准在 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": "llm-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大模型基准论文？
共 544 篇，按年份分别是 2023 年 2 篇、2024 年 61 篇、2025 年 174 篇、2026 年 307 篇。
NeurIPS 2026 有多少篇大模型基准论文？
NeurIPS 2026 有 138 篇大模型基准论文，上一届是 92 篇。
ICLR 2026 有多少篇大模型基准论文？
ICLR 2026 有 103 篇大模型基准论文，上一届是 50 篇。
ICML 2026 有多少篇大模型基准论文？
ICML 2026 有 60 篇大模型基准论文，上一届是 24 篇。
CVPR 2026 有多少篇大模型基准论文？
CVPR 2026 有 6 篇大模型基准论文，上一届是 8 篇。
哪个会议的大模型基准论文最多？
NeurIPS，2023–2026 年共 256 篇。

「评测与基准」下的其他主题