深度学习 · large language models论文趋势 2023–2026

ICML/NeurIPS 深度学习方向：large_language_models

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 2,334 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

深度学习 · large language models论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
20234150730488
2024298000298
20250134040417
2026001,13101,131

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

全部深度学习 · large language models论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 2,334 篇，第 31 / 47 页

  1. ICML 2025The Berkeley Function Calling Leaderboard (BFCL): From Tool Use to Agentic Evaluation of Large Language Models
  2. ICML 2025The Best of Both Worlds: Bridging Quality and Diversity in Data Selection with Bipartite Graph
  3. ICML 2025The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
  4. ICML 2025The Emperor's New Clothes in Benchmarking? A Rigorous Examination of Mitigation Strategies for LLM Benchmark Data Contamination
  5. ICML 2025The Energy Loss Phenomenon in RLHF: A New Perspective on Mitigating Reward Hacking
  6. ICML 2025The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence
  7. ICML 2025The Hidden Dimensions of LLM Alignment: A Multi-Dimensional Analysis of Orthogonal Safety Directions
  8. ICML 2025The Missing Alignment Link of In-context Learning on Sequences
  9. ICML 2025The Surprising Effectiveness of Test-Time Training for Few-Shot Learning
  10. ICML 2025The Synergy of LLMs & RL Unlocks Offline Learning of Generalizable Language-Conditioned Policies with Low-fidelity Data
  11. ICML 2025Think Smarter not Harder: Adaptive Reasoning with Inference Aware Optimization
  12. ICML 2025Thinking LLMs: General Instruction Following with Thought Generation
  13. ICML 2025To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models
  14. ICML 2025Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning
  15. ICML 2025Token Cleaning: Fine-Grained Data Selection for LLM Supervised Fine-Tuning
  16. ICML 2025Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models
  17. ICML 2025TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation
  18. ICML 2025Towards Cost-Effective Reward Guided Text Generation
  19. ICML 2025Towards Lifelong Model Editing via Simulating Ideal Editor
  20. ICML 2025Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
  21. ICML 2025Training a Generally Curious Agent
  22. ICML 2025Training Software Engineering Agents and Verifiers with SWE-Gym
  23. ICML 2025TruthFlow: Truthful LLM Generation via Representation Flow Correction
  24. ICML 2025TuCo: Measuring the Contribution of Fine-Tuning to Individual Responses of LLMs
  25. ICML 2025Tuning LLM Judge Design Decisions for 1/1000 of the Cost
  26. ICML 2025UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning
  27. ICML 2025Unbiased Evaluation of Large Language Models from a Causal Perspective
  28. ICML 2025Understanding Bias Reinforcement in LLM Agents Debate
  29. ICML 2025Understanding Chain-of-Thought in LLMs through Information Theory
  30. ICML 2025Understanding Multimodal LLMs Under Distribution Shifts: An Information-Theoretic Approach
  31. ICML 2025Understanding Synthetic Context Extension via Retrieval Heads
  32. ICML 2025Understanding the Skill Gap in Recurrent Language Models: The Role of the Gather-and-Aggregate Mechanism
  33. ICML 2025Unlocking Post-hoc Dataset Inference with Synthetic Data
  34. ICML 2025Unnatural Languages Are Not Bugs but Features for LLMs
  35. ICML 2025VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data
  36. ICML 2025VinePPO: Refining Credit Assignment in RL Training of LLMs
  37. ICML 2025Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning
  38. ICML 2025What Do Learning Dynamics Reveal About Generalization in LLM Mathematical Reasoning?
  39. ICML 2025What Limits Bidirectional Model's Generative Capabilities? A Uni-Bi-Directional Mixture-of-Expert Method For Bidirectional Fine-tuning
  40. ICML 2025What Makes In-context Learning Effective for Mathematical Reasoning
  41. ICML 2025When Bad Data Leads to Good Models
  42. ICML 2025Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?
  43. ICML 2025Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas
  44. ICML 2025WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs
  45. ICML 2025XAttention: Block Sparse Attention with Antidiagonal Scoring
  46. ICML 2025xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference
  47. ICML 2025ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
  48. ICML 2025µnit Scaling: Simple and Scalable FP8 LLM Training
  49. NeurIPS 2024$\beta$-DPO: Direct Preference Optimization with Dynamic $\beta$
  50. NeurIPS 2024$\textit{Trans-LoRA}$: towards data-free Transferable Parameter Efficient Finetuning

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 查深度学习 · large language models在 ICML 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": "dl-large-language-models"}

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深度学习 · large language models论文？
共 2,334 篇，按年份分别是 2023 年 488 篇、2024 年 298 篇、2025 年 417 篇、2026 年 1,131 篇。
NeurIPS 2024 有多少篇深度学习 · large language models论文？
NeurIPS 2024 有 298 篇深度学习 · large language models论文，上一届是 415 篇。
ICLR 2025 有多少篇深度学习 · large language models论文？
ICLR 2025 有 13 篇深度学习 · large language models论文。
ICML 2026 有多少篇深度学习 · large language models论文？
ICML 2026 有 1,131 篇深度学习 · large language models论文，上一届是 404 篇。
哪个会议的深度学习 · large language models论文最多？
ICML，2023–2026 年共 1,608 篇。

「深度学习」下的其他主题