人工评测论文趋势 2023–2026

主要结论

• 人工评测论文从 2023 年的 33 篇增长到 2026 年的 108 篇，三年间扩大到约 3.3 倍。
• NeurIPS 是发人工评测论文最多的会议，2023–2026 年共 90 篇，占四个会议总量的 33%。
• 交叉方向里涨得最快的是数据集与基准：同时属于两个主题的论文从 2023 年的 3 篇增至 2026 年的 31 篇。
• 在「评测与基准」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的大模型基准是 154 倍，人工评测是 3.3 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 2026Ad-Hoc Teamwork from Human Demonstrations
• ICML 2026SWE-IF: Aligning Code Evaluation with Human Preference
• ICLR 2026Human-LLM Collaborative Feature Engineering for Tabular Data
• CVPR 2026Bridging Human Evaluation to Infrared and Visible Image Fusion
• NeurIPS 2025Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning
• ICML 2025Position: When Incentives Backfire, Data Stops Being Human
• ICLR 2025Eliciting Human Preferences with Language Models
• CVPR 2025DynPose: Largely Improving the Efficiency of Human Pose Estimation by a Simple Dynamic Framework
• NeurIPS 2024DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA
• ICML 2024A Fine-grained Analysis of Fitted Q-evaluation: Beyond Parametric Models
• ICLR 2024Human Feedback is not Gold Standard
• CVPR 2024LocLLM: Exploiting Generalizable Human Keypoint Localization via Large Language Model
• NeurIPS 2023Privacy Assessment on Reconstructed Images: Are Existing Evaluation Metrics Faithful to Human Perception?
• ICML 2023Regions of Reliability in the Evaluation of Multivariate Probabilistic Forecasts
• ICLR 2023Beyond calibration: estimating the grouping loss of modern neural networks
• CVPR 20233D Human Keypoints Estimation From Point Clouds in the Wild Without Human Labels
• NeurIPS 2026AI Construct Lexis: An Ontology of the Hidden Assumptions in AI Evaluation
• ICML 2026Rethinking Evaluation Paradigms in IBP-based Certified Training
• ICLR 2026Multi-turn Evaluation of Anthropomorphic Behaviours in Large Language Models
• CVPR 2026Ground Reaction Inertial Poser Physics-based Human Motion Capture from Sparse

全部人工评测论文

共 60 篇，第 1 / 2 页

1. CVPR 2025DynPose: Largely Improving the Efficiency of Human Pose Estimation by a Simple Dynamic Framework
2. CVPR 2025Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects
3. CVPR 2025Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching
4. CVPR 2025Image Quality Assessment: From Human to Machine Preference
5. CVPR 2025MixerMDM: Learnable Composition of Human Motion Diffusion Models
6. CVPR 2025Nonisotropic Gaussian Diffusion for Realistic 3D Human Motion Prediction
7. CVPR 2025PI-HMR: Towards Robust In-bed Temporal Human Shape Reconstruction with Contact Pressure Sensing
8. CVPR 2025ProbPose: A Probabilistic Approach to 2D Human Pose Estimation
9. CVPR 2025RePerformer: Immersive Human-centric Volumetric Videos from Playback to Photoreal Reperformance
10. CVPR 2025Rethinking Diffusion for Text-Driven Human Motion Generation: Redundant Representations, Evaluation, and Masked Autoregression
11. CVPR 2025StickMotion: Generating 3D Human Motions by Drawing a Stickman
12. CVPR 2025Towards Scalable Human-aligned Benchmark for Text-guided Image Editing
13. CVPR 2025Video-Bench: Human-Aligned Video Generation Benchmark
14. ICLR 2025An Empirical Analysis of Uncertainty in Large Language Model Evaluations
15. ICLR 2025Beyond correlation: The impact of human uncertainty in measuring the effectiveness of automatic evaluation and LLM-as-a-judge
16. ICLR 2025Consistency Checks for Language Model Forecasters
17. ICLR 2025DreamBench++: A Human-Aligned Benchmark for Personalized Image Generation
18. ICLR 2025Eliciting Human Preferences with Language Models
19. ICLR 2025Human Simulacra: Benchmarking the Personification of Large Language Models
20. ICLR 2025Human-Aligned Chess With a Bit of Search
21. ICLR 2025Inverse Constitutional AI: Compressing Preferences into Principles
22. ICLR 2025Language Models Learn to Mislead Humans via RLHF
23. ICLR 2025Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation
24. ICLR 2025On Speeding Up Language Model Evaluation
25. ICLR 2025Realistic Evaluation of Deep Partial-Label Learning Algorithms
26. ICLR 2025Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?
27. ICLR 2025RevisEval: Improving LLM-as-a-Judge via Response-Adapted References
28. ICLR 2025Revisiting text-to-image evaluation with Gecko: on metrics, prompts, and human rating
29. ICLR 2025Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreement
30. ICML 2025Accelerating Unbiased LLM Evaluation via Synthetic Feedback
31. ICML 2025AutoEval Done Right: Using Synthetic Data for Model Evaluation
32. ICML 2025CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective
33. ICML 2025GRADEO: Towards Human-Like Evaluation for Text-to-Video Generation via Multi-Step Reasoning
34. ICML 2025MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models
35. ICML 2025MCU: An Evaluation Framework for Open-Ended Game Agents
36. ICML 2025Position: AI Competitions Provide the Gold Standard for Empirical Rigor in GenAI Evaluation
37. ICML 2025Position: AI Evaluation Should Learn from How We Test Humans
38. ICML 2025Position: AI Should Not Be An Imitation Game: Centaur Evaluations
39. ICML 2025Position: Democratic AI is Possible. The Democracy Levels Framework Shows How It Might Work.
40. ICML 2025Position: Human Baselines in Model Evaluations Need Rigor and Transparency (With Recommendations & Reporting Checklist)
41. ICML 2025Position: Stop treating `AGI' as the north-star goal of AI research
42. ICML 2025Position: When Incentives Backfire, Data Stops Being Human
43. ICML 2025Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing
44. NeurIPS 2025A Principled Path to Fitted Distributional Evaluation
45. NeurIPS 2025AGENTIF: Benchmarking Large Language Models Instruction Following Ability in Agentic Scenarios
46. NeurIPS 2025AI Debate Aids Assessment of Controversial Claims
47. NeurIPS 2025Beyond the Surface: Enhancing LLM-as-a-Judge Alignment with Human via Internal Representations
48. NeurIPS 2025Can Large Language Models Master Complex Card Games?
49. NeurIPS 2025Can LLMs Outshine Conventional Recommenders? A Comparative Evaluation
50. NeurIPS 2025Clean First, Align Later: Benchmarking Preference Data Cleaning for Reliable LLM Alignment

常见问题

2023–2026 年四大会议一共有多少篇人工评测论文？

共 271 篇，按年份分别是 2023 年 33 篇、2024 年 70 篇、2025 年 60 篇、2026 年 108 篇。

NeurIPS 2026 有多少篇人工评测论文？

NeurIPS 2026 有 39 篇人工评测论文，上一届是 17 篇。

ICLR 2026 有多少篇人工评测论文？

ICLR 2026 有 29 篇人工评测论文，上一届是 16 篇。

ICML 2026 有多少篇人工评测论文？

ICML 2026 有 21 篇人工评测论文，上一届是 14 篇。

CVPR 2026 有多少篇人工评测论文？

CVPR 2026 有 19 篇人工评测论文，上一届是 13 篇。

哪个会议的人工评测论文最多？

NeurIPS，2023–2026 年共 90 篇。