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 307 篇，第 1 / 7 页

1. CVPR 20263D-VCD Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive
2. CVPR 20263DrawAgent Teaching LLM to Draw in 3D with Early Contrastive
3. CVPR 2026ArtLLM Generating Articulated Assets via 3D LLM
4. CVPR 2026DualMirage Hunting Stealthy Multimodal LLM Agents via CAPTCHAs with Contour
5. CVPR 2026Prototype-as-Prompt Multimodal Sentiment Prototypes Endowing Large Language Models the Capability
6. CVPR 2026RAAS LLM Agentic System Architecture Search with GRPO
7. ICLR 2026<SO$G_k$>: One LLM Token for Explicit Graph Structural Understanding
8. ICLR 2026Activation Steering for LLM Alignment via a Unified ODE-Based Framework
9. ICLR 2026Adaptive Thinking: Large Language Models Know When to Think in Latent Space
10. ICLR 2026ASCIIEval: Benchmarking Models' Visual Perception in Text Strings via ASCII Art
11. ICLR 2026AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism
12. ICLR 2026Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data
13. ICLR 2026Calibrating Verbalized Confidence with Self-Generated Distractors
14. ICLR 2026Cancer-Myth: Evaluating Large Language Models on Patient Questions with False Presuppositions
15. ICLR 2026Capability-Based Scaling Trends for LLM-Based Red-Teaming
16. ICLR 2026CMPhysBench: A Benchmark for Evaluating Large Language Models in Condensed Matter Physics
17. ICLR 2026Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Training
18. ICLR 2026Count Counts: Motivating Exploration in LLM Reasoning with Count-based Intrinsic Rewards
19. ICLR 2026Credit-Budgeted ICPC-Style Coding: When LLM Agents Must Pay for Every Decision
20. ICLR 2026DASH: Deterministic Attention Scheduling for High-throughput Reproducible LLM Training
21. ICLR 2026Data Selection for LLM Alignment Using Fine-Grained Preferences
22. ICLR 2026DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents
23. ICLR 2026Detecting Data Contamination in LLMs via In-Context Learning
24. ICLR 2026Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement
25. ICLR 2026Discovering Novel LLM Experts via Task-Capability Coevolution
26. ICLR 2026Downgrade to Upgrade: Optimizer Simplification Enhances Robustness in LLM Unlearning
27. ICLR 2026Dr.LLM: Dynamic Layer Routing in LLMs
28. ICLR 2026EditBench: Evaluating LLM Abilities to Perform Real-World Instructed Code Edits
29. ICLR 2026Empowering LLM Tool Invocation with Tool-call Reward Model
30. ICLR 2026Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection
31. ICLR 2026Epistemic Uncertainty Quantification To Improve Decisions From Black-Box Models
32. ICLR 2026Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesis
33. ICLR 2026ExpGuard: LLM Content Moderation in Specialized Domains
34. ICLR 2026Fast-dLLM v2: Efficient Block-Diffusion LLM
35. ICLR 2026From Static Benchmarks to Dynamic Protocol: Agent-Centric Text Anomaly Detection for Evaluating LLM Reasoning
36. ICLR 2026FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction
37. ICLR 2026Go-Browse: Training Web Agents with Structured Exploration
38. ICLR 2026Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
39. ICLR 2026How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability
40. ICLR 2026How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining
41. ICLR 2026How Stable is the Next Token? A Geometric View of LLM Prediction Stability
42. ICLR 2026Improving LLM-based Global Optimization with Search Space Partitioning
43. ICLR 2026In Agents We Trust, but Who Do Agents Trust? Latent Preferences Steer LLM Generations
44. ICLR 2026Jet Expansions: Restructuring LLM Computation for Model Inspection
45. ICLR 2026KL-Regularized Reinforcement Learning is Designed to Mode Collapse
46. ICLR 2026KV-Cache Transform Coding for Compact Storage in LLM Inference
47. ICLR 2026LLM as an Algorithmist: Enhancing Anomaly Detectors via Programmatic Synthesis
48. ICLR 2026LLM Pretraining with Continuous Concepts
49. ICLR 2026LLM Unlearning with LLM Beliefs
50. ICLR 2026LLM-as-a-Prophet: Understanding Predictive Intelligence with Prophet Arena

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。