Human Evaluation papers, 2023–2026

Key findings

• Human Evaluation went from 33 papers in 2023 to 108 in 2026, roughly 3.3x in three years.
• NeurIPS publishes the most Human Evaluation work: 90 papers from 2023 to 2026, 33% of the four-venue total.
• The fastest-growing overlap is with Datasets and Benchmarks: papers tagged with both rose from 3 in 2023 to 31 in 2026.
• Among the 5 topics under Evaluation and Benchmarks, it ranks #3 by 2026 paper count; the fastest grower in the group, LLM Benchmarks, is up 154x versus 3.3x here.

Representative papers

Spread across years and venues so every edition shows up

• 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

All Human Evaluation papers

271 papers, page 3 of 6

101. NeurIPS 2026Select Smarter, Not More? Prompt-Aware Evaluation Scheduling with Submodular Guarantees
102. NeurIPS 2026SR-Prominence: A Crowdsourced Protocol and Dataset Suite for Perceptually-Weighted Super-Resolution Artifact Evaluation
103. NeurIPS 2026The Deadline Effect: Identifying and Correcting Temporal Bias in Human Evaluation
104. NeurIPS 2026Verifier Choice is a Benchmark Design Variable: Auditing Structural Counting Evaluation in Text-to-Image Models
105. NeurIPS 2026VIPBench: A Human-Aligned Benchmark for Voice Identity Perception in the Age of Voice Cloning
106. NeurIPS 2026VTBench: Disentangled and Human-Aligned Evaluation for Image-Based Virtual Try-on
107. NeurIPS 2026When Are Predictions Enough? An Evaluation Protocol for Frozen Expert Composition
108. NeurIPS 2026WorldReasonBench: Human-Aligned Stress Testing of Video Generators as Future World-State Predictors
109. CVPR 2025DynPose: Largely Improving the Efficiency of Human Pose Estimation by a Simple Dynamic Framework
110. CVPR 2025Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects
111. CVPR 2025Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching
112. CVPR 2025Image Quality Assessment: From Human to Machine Preference
113. CVPR 2025MixerMDM: Learnable Composition of Human Motion Diffusion Models
114. CVPR 2025Nonisotropic Gaussian Diffusion for Realistic 3D Human Motion Prediction
115. CVPR 2025PI-HMR: Towards Robust In-bed Temporal Human Shape Reconstruction with Contact Pressure Sensing
116. CVPR 2025ProbPose: A Probabilistic Approach to 2D Human Pose Estimation
117. CVPR 2025RePerformer: Immersive Human-centric Volumetric Videos from Playback to Photoreal Reperformance
118. CVPR 2025Rethinking Diffusion for Text-Driven Human Motion Generation: Redundant Representations, Evaluation, and Masked Autoregression
119. CVPR 2025StickMotion: Generating 3D Human Motions by Drawing a Stickman
120. CVPR 2025Towards Scalable Human-aligned Benchmark for Text-guided Image Editing
121. CVPR 2025Video-Bench: Human-Aligned Video Generation Benchmark
122. ICLR 2025An Empirical Analysis of Uncertainty in Large Language Model Evaluations
123. ICLR 2025Beyond correlation: The impact of human uncertainty in measuring the effectiveness of automatic evaluation and LLM-as-a-judge
124. ICLR 2025Consistency Checks for Language Model Forecasters
125. ICLR 2025DreamBench++: A Human-Aligned Benchmark for Personalized Image Generation
126. ICLR 2025Eliciting Human Preferences with Language Models
127. ICLR 2025Human Simulacra: Benchmarking the Personification of Large Language Models
128. ICLR 2025Human-Aligned Chess With a Bit of Search
129. ICLR 2025Inverse Constitutional AI: Compressing Preferences into Principles
130. ICLR 2025Language Models Learn to Mislead Humans via RLHF
131. ICLR 2025Large Language Models Meet Symbolic Provers for Logical Reasoning Evaluation
132. ICLR 2025On Speeding Up Language Model Evaluation
133. ICLR 2025Realistic Evaluation of Deep Partial-Label Learning Algorithms
134. ICLR 2025Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?
135. ICLR 2025RevisEval: Improving LLM-as-a-Judge via Response-Adapted References
136. ICLR 2025Revisiting text-to-image evaluation with Gecko: on metrics, prompts, and human rating
137. ICLR 2025Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreement
138. ICML 2025Accelerating Unbiased LLM Evaluation via Synthetic Feedback
139. ICML 2025AutoEval Done Right: Using Synthetic Data for Model Evaluation
140. ICML 2025CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective
141. ICML 2025GRADEO: Towards Human-Like Evaluation for Text-to-Video Generation via Multi-Step Reasoning
142. ICML 2025MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models
143. ICML 2025MCU: An Evaluation Framework for Open-Ended Game Agents
144. ICML 2025Position: AI Competitions Provide the Gold Standard for Empirical Rigor in GenAI Evaluation
145. ICML 2025Position: AI Evaluation Should Learn from How We Test Humans
146. ICML 2025Position: AI Should Not Be An Imitation Game: Centaur Evaluations
147. ICML 2025Position: Democratic AI is Possible. The Democracy Levels Framework Shows How It Might Work.
148. ICML 2025Position: Human Baselines in Model Evaluations Need Rigor and Transparency (With Recommendations & Reporting Checklist)
149. ICML 2025Position: Stop treating `AGI' as the north-star goal of AI research
150. ICML 2025Position: When Incentives Backfire, Data Stops Being Human

FAQ

How many Human Evaluation papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

271 in total: 33 in 2023, 70 in 2024, 60 in 2025, 108 in 2026.

How many Human Evaluation papers are at NeurIPS 2026?

NeurIPS 2026 has 39 Human Evaluation papers, versus 17 the year before.

How many Human Evaluation papers are at ICLR 2026?

ICLR 2026 has 29 Human Evaluation papers, versus 16 the year before.

How many Human Evaluation papers are at ICML 2026?

ICML 2026 has 21 Human Evaluation papers, versus 14 the year before.

How many Human Evaluation papers are at CVPR 2026?

CVPR 2026 has 19 Human Evaluation papers, versus 13 the year before.

Which conference publishes the most Human Evaluation papers?

NeurIPS, with 90 papers from 2023 to 2026.