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 256 篇，第 4 / 6 页

  1. NeurIPS 2025BrainEC-LLM: Brain Effective Connectivity Estimation by Multiscale Mixing LLM
  2. NeurIPS 2025C-SEO Bench: Does Conversational SEO Work?
  3. NeurIPS 2025Can Dependencies Induced by LLM-Agent Workflows Be Trusted?
  4. NeurIPS 2025Causal LLM Routing: End-to-End Regret Minimization from Observational Data
  5. NeurIPS 2025ComPO: Preference Alignment via Comparison Oracles
  6. NeurIPS 2025ConnectomeBench: Can LLMs proofread the connectome?
  7. NeurIPS 2025Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning
  8. NeurIPS 2025DartQuant: Efficient Rotational Distribution Calibration for LLM Quantization
  9. NeurIPS 2025Dependency Matters: Enhancing LLM Reasoning with Explicit Knowledge Grounding
  10. NeurIPS 2025DISC: Dynamic Decomposition Improves LLM Inference Scaling
  11. NeurIPS 2025Distribution-Aligned Decoding for Efficient LLM Task Adaptation
  12. NeurIPS 2025Do Language Models Use Their Depth Efficiently?
  13. NeurIPS 2025DP-LLM: Runtime Model Adaptation with Dynamic Layer-wise Precision Assignment
  14. NeurIPS 2025EffiBench-X: A Multi-Language Benchmark for Measuring Efficiency of LLM-Generated Code
  15. NeurIPS 2025Efficient Large Language Model Inference with Neural Block Linearization
  16. NeurIPS 2025Enhancing Multilingual LLM Pretraining with Model-Based Data Selection
  17. NeurIPS 2025Evaluating LLM-contaminated Crowdsourcing Data Without Ground Truth
  18. NeurIPS 2025Extrapolation by Association: Length Generalization Transfer In Transformers
  19. NeurIPS 2025Factorio Learning Environment
  20. NeurIPS 2025FFN Fusion: Rethinking Sequential Computation in Large Language Models
  21. NeurIPS 2025FreshStack: Building Realistic Benchmarks for Evaluating Retrieval on Technical Documents
  22. NeurIPS 2025GPAS: Accelerating Convergence of LLM Pretraining via Gradient-Preserving Activation Scaling
  23. NeurIPS 2025GPO: Learning from Critical Steps to Improve LLM Reasoning
  24. NeurIPS 2025ICPC-Eval: Probing the Frontiers of LLM Reasoning with Competitive Programming Contests
  25. NeurIPS 2025IR-OptSet: An Optimization-Sensitive Dataset for Advancing LLM-Based IR Optimizer
  26. NeurIPS 2025Learning Grouped Lattice Vector Quantizers for Low-Bit LLM Compression
  27. NeurIPS 2025Less is More: Improving LLM Alignment via Preference Data Selection
  28. NeurIPS 2025Let the LLM Stick to Its Strengths: Learning to Route Economical LLM
  29. NeurIPS 2025LLM Layers Immediately Correct Each Other
  30. NeurIPS 2025LLM Meeting Decision Trees on Tabular Data
  31. NeurIPS 2025LLM-DAMVC: A Large Language Model Assisted Dynamic Agent for Multi-View Clustering
  32. NeurIPS 2025LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models
  33. NeurIPS 2025Lookahead Routing for Large Language Models
  34. NeurIPS 2025LTD-Bench: Evaluating Large Language Models by Letting Them Draw
  35. NeurIPS 2025Lua-LLM: Learning Unstructured-Sparsity Allocation for Large Language Models
  36. NeurIPS 2025Masked Gated Linear Unit
  37. NeurIPS 2025MESS+: Dynamically Learned Inference-Time LLM Routing in Model Zoos with Service Level Guarantees
  38. NeurIPS 2025metaTextGrad: Automatically optimizing language model optimizers
  39. NeurIPS 2025Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning
  40. NeurIPS 2025MODEL SHAPLEY: Find Your Ideal Parameter Player via One Gradient Backpropagation
  41. NeurIPS 2025OmniBench: Towards The Future of Universal Omni-Language Models
  42. NeurIPS 2025OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and Metrics
  43. NeurIPS 2025PARALLELPROMPT: Extracting Parallelism from Large Language Model Queries
  44. NeurIPS 2025Polar Sparsity: High Throughput Batched LLM Inferencing with Scalable Contextual Sparsity
  45. NeurIPS 2025Predicting Empirical AI Research Outcomes with Language Models
  46. NeurIPS 2025Preserving LLM Capabilities through Calibration Data Curation: From Analysis to Optimization
  47. NeurIPS 2025Q-Palette: Fractional-Bit Quantizers Toward Optimal Bit Allocation for Efficient LLM Deployment
  48. NeurIPS 2025QuadEnhancer: Leveraging Quadratic Transformations to Enhance Deep Neural Networks
  49. NeurIPS 2025REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websites
  50. NeurIPS 2025Reasoning Is Not a Race: When Stopping Early Beats Going Deeper

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