大模型基准论文趋势 2023–2026

主要结论

• 大模型基准论文从 2023 年的 2 篇增长到 2026 年的 307 篇，三年间扩大到约 154 倍。
• NeurIPS 是发大模型基准论文最多的会议，2023–2026 年共 256 篇，占四个会议总量的 47%。
• 交叉方向里涨得最快的是AI 智能体：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 116 篇。
• 在「评测与基准」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的模型即裁判是 27 倍，大模型基准是 154 倍。

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

• NeurIPS 2026A Free Lunch in LLM Compression: Revisiting Retraining after Pruning
• ICML 2026Controlled LLM Training on Spectral Sphere
• ICLR 2026Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
• CVPR 20263DrawAgent Teaching LLM to Draw in 3D with Early Contrastive
• NeurIPS 2025ACCO: Accumulate While You Communicate for Communication-Overlapped Sharded LLM Training
• ICML 2025Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM’s Reasoning Capability
• ICLR 2025Catastrophic Failure of LLM Unlearning via Quantization
• CVPR 2025Period-LLM: Extending the Periodic Capability of Multimodal Large Language Model
• NeurIPS 2024Large Language Models' Expert-level Global History Knowledge Benchmark (HiST-LLM)
• ICML 2024Characterizing ResNet's Universal Approximation Capability
• ICLR 2024Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactuals
• CVPR 2024Koala: Key Frame-Conditioned Long Video-LLM
• NeurIPS 2023Quantizable Transformers: Removing Outliers by Helping Attention Heads Do Nothing
• NeurIPS 2026Accelerating LLM Pre-Training through Flat-Direction Dynamics Enhancement
• ICML 2026LLM Priors for ERM over Programs
• ICLR 2026How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining
• CVPR 20263D-VCD Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive
• NeurIPS 2025BrainEC-LLM: Brain Effective Connectivity Estimation by Multiscale Mixing LLM
• ICML 2025Position: Medical Large Language Model Benchmarks Should Prioritize Construct Validity
• ICLR 2025Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

全部大模型基准论文

共 61 篇，第 1 / 2 页

1. CVPR 2024Honeybee: Locality-enhanced Projector for Multimodal LLM
2. CVPR 2024Koala: Key Frame-Conditioned Long Video-LLM
3. CVPR 2024Self-correcting LLM-controlled Diffusion Models
4. ICLR 2024BEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks
5. ICLR 2024Can LLM-Generated Misinformation Be Detected?
6. ICLR 2024ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate
7. ICLR 2024Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactuals
8. ICLR 2024LLM-Assisted Code Cleaning For Training Accurate Code Generators
9. ICLR 2024LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset
10. ICLR 2024PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization
11. ICLR 2024PB-LLM: Partially Binarized Large Language Models
12. ICLR 2024Prometheus: Inducing Fine-Grained Evaluation Capability in Language Models
13. ICLR 2024SKILL-MIX: a Flexible and Expandable Family of Evaluations for AI Models
14. ICLR 2024Steve-Eye: Equipping LLM-based Embodied Agents with Visual Perception in Open Worlds
15. ICLR 2024The LLM Surgeon
16. ICLR 2024To the Cutoff... and Beyond? A Longitudinal Perspective on LLM Data Contamination
17. ICML 2024An LLM Compiler for Parallel Function Calling
18. ICML 2024Any-Precision LLM: Low-Cost Deployment of Multiple, Different-Sized LLMs
19. ICML 2024Auto-Encoding Morph-Tokens for Multimodal LLM
20. ICML 2024Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
21. ICML 2024Characterizing ResNet's Universal Approximation Capability
22. ICML 2024CLLMs: Consistency Large Language Models
23. ICML 2024Embodied CoT Distillation From LLM To Off-the-shelf Agents
24. ICML 2024ExCP: Extreme LLM Checkpoint Compression via Weight-Momentum Joint Shrinking
25. ICML 2024Executable Code Actions Elicit Better LLM Agents
26. ICML 2024Exploring the LLM Journey from Cognition to Expression with Linear Representations
27. ICML 2024Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study
28. ICML 2024LLM Maybe LongLM: SelfExtend LLM Context Window Without Tuning
29. ICML 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
30. ICML 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
31. ICML 2024MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark
32. ICML 2024MuxServe: Flexible Spatial-Temporal Multiplexing for Multiple LLM Serving
33. ICML 2024NExT-GPT: Any-to-Any Multimodal LLM
34. ICML 2024Position: Fundamental Limitations of LLM Censorship Necessitate New Approaches
35. ICML 2024Position: Key Claims in LLM Research Have a Long Tail of Footnotes
36. ICML 2024Reason for Future, Act for Now: A Principled Architecture for Autonomous LLM Agents
37. ICML 2024SparQ Attention: Bandwidth-Efficient LLM Inference
38. NeurIPS 2024Cascade Speculative Drafting for Even Faster LLM Inference
39. NeurIPS 2024DISP-LLM: Dimension-Independent Structural Pruning for Large Language Models
40. NeurIPS 2024Distributional Preference Alignment of LLMs via Optimal Transport
41. NeurIPS 2024EHRNoteQA: An LLM Benchmark for Real-World Clinical Practice Using Discharge Summaries
42. NeurIPS 2024Enhancing LLM’s Cognition via Structurization
43. NeurIPS 2024Evaluating Copyright Takedown Methods for Language Models
44. NeurIPS 2024Exploiting LLM Quantization
45. NeurIPS 2024Gorilla: Large Language Model Connected with Massive APIs
46. NeurIPS 2024HYDRA: Model Factorization Framework for Black-Box LLM Personalization
47. NeurIPS 2024IaC-Eval: A Code Generation Benchmark for Cloud Infrastructure-as-Code Programs
48. NeurIPS 2024Large Language Models' Expert-level Global History Knowledge Benchmark (HiST-LLM)
49. NeurIPS 2024LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language
50. NeurIPS 2024LLM-AutoDA: Large Language Model-Driven Automatic Data Augmentation for Long-tailed Problems

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。