Weak Supervision papers, 2023–2026

Key findings

• Weak Supervision went from 34 papers in 2023 to 45 in 2026, roughly 1.3x in three years.
• NeurIPS publishes the most Weak Supervision work: 42 papers from 2023 to 2026, 31% of the four-venue total.
• The fastest-growing overlap is with Alignment and Safety: papers tagged with both rose from 0 in 2023 to 7 in 2026.
• Among the 5 topics under Data-Centric AI, it ranks #2 by 2026 paper count; the fastest grower in the group, Synthetic Data Generation, is up 1.3x versus 1.3x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026Affine-invariant Cubic Newton with Weak Learners
• ICML 2026Weak-to-Strong Generalization via Bregman Bias–Variance Decomposition
• ICLR 2026Weak-to-Strong Generalization with Failure Trajectories
• CVPR 2026GEM-TFL Bridging Weak and Full Supervision for Forgery Localization through
• NeurIPS 2025Disentangled Cross-Modal Representation Learning with Enhanced Mutual Supervision
• ICML 2025How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction
• ICLR 2025LASER: A Neuro-Symbolic Framework for Learning Spatio-Temporal Scene Graphs with Weak Supervision
• CVPR 2025Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer
• NeurIPS 2024Metrizing Weak Convergence with Maximum Mean Discrepancies
• ICML 2024A General Framework for Learning from Weak Supervision
• ICLR 2024Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks
• CVPR 2024Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision
• NeurIPS 2023Latent Graph Inference with Limited Supervision
• ICML 2023AdaBoost is not an Optimal Weak to Strong Learner
• ICLR 2023Generative Modeling Helps Weak Supervision (and Vice Versa)
• CVPR 2023DPF: Learning Dense Prediction Fields With Weak Supervision
• NeurIPS 2026Clustering with Weak Distance Oracles
• ICML 2026Learning Permutation from Structure Without Supervision
• ICLR 2026Language-guided Open-world Video Anomaly Detection under Weak Supervision
• CVPR 2026Generalizable Video Quality Assessment via Weak-to-Strong Learning

All Weak Supervision papers

136 papers, page 2 of 3

51. ICLR 2025Iterative Label Refinement Matters More than Preference Optimization under Weak Supervision
52. ICLR 2025LASER: A Neuro-Symbolic Framework for Learning Spatio-Temporal Scene Graphs with Weak Supervision
53. ICLR 2025Learning from weak labelers as constraints
54. ICLR 2025Provable weak-to-strong generalization via benign overfitting
55. ICLR 2025SMI-Editor: Edit-based SMILES Language Model with Fragment-level Supervision
56. ICLR 2025Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization
57. ICLR 2025Weak to Strong Generalization for Large Language Models with Multi-capabilities
58. ICLR 2025Your Weak LLM is Secretly a Strong Teacher for Alignment
59. ICML 2025How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction
60. ICML 2025Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions
61. NeurIPS 2025Advancing Machine-Generated Text Detection from an Easy to Hard Supervision Perspective
62. NeurIPS 2025Disentangled Cross-Modal Representation Learning with Enhanced Mutual Supervision
63. NeurIPS 2025Disentangling Latent Shifts of In-Context Learning with Weak Supervision
64. NeurIPS 2025Escaping Collapse: The Strength of Weak Data for Large Language Model Training
65. NeurIPS 2025From Linear to Nonlinear: Provable Weak-to-Strong Generalization through Feature Learning
66. NeurIPS 2025Learning from Interval Targets
67. NeurIPS 2025MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering
68. NeurIPS 2025On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective
69. NeurIPS 2025Revisiting Agnostic Boosting
70. NeurIPS 2025Revisiting End-to-End Learning with Slide-level Supervision in Computational Pathology
71. NeurIPS 2025Robust Label Proportions Learning
72. NeurIPS 2025SEGA: Shaping Semantic Geometry for Robust Hashing under Noisy Supervision
73. NeurIPS 2025Statistical Analysis of an Adversarial Bayesian Weak Supervision Method
74. NeurIPS 2025Weak-to-Strong Generalization under Distribution Shifts
75. NeurIPS 2025Weaver: Shrinking the Generation-Verification Gap by Scaling Compute for Verification
76. CVPR 2024Atom-Level Optical Chemical Structure Recognition with Limited Supervision
77. CVPR 2024Enhancing Visual Continual Learning with Language-Guided Supervision
78. CVPR 2024Improving Distant 3D Object Detection Using 2D Box Supervision
79. CVPR 2024Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision
80. CVPR 2024MS-DETR: Efficient DETR Training with Mixed Supervision
81. CVPR 2024PointOBB: Learning Oriented Object Detection via Single Point Supervision
82. CVPR 2024SEAS: ShapE-Aligned Supervision for Person Re-Identification
83. ICLR 2024Adversarial Supervision Makes Layout-to-Image Diffusion Models Thrive
84. ICLR 2024Towards a statistical theory of data selection under weak supervision
85. ICLR 2024Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks
86. ICML 2024A General Framework for Learning from Weak Supervision
87. ICML 2024Fast Co-Training under Weak Dependence via Stream-Based Active Learning
88. ICML 2024On the Effectiveness of Supervision in Asymmetric Non-Contrastive Learning
89. ICML 2024ULAREF: A Unified Label Refinement Framework for Learning with Inaccurate Supervision
90. ICML 2024Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
91. ICML 2024WISER: Weak Supervision and Supervised Representation Learning to Improve Drug Response Prediction in Cancer
92. NeurIPS 2024Intrinsic Self-Supervision for Data Quality Audits
93. NeurIPS 2024Metrizing Weak Convergence with Maximum Mean Discrepancies
94. NeurIPS 2024On scalable oversight with weak LLMs judging strong LLMs
95. NeurIPS 2024Quantifying the Gain in Weak-to-Strong Generalization
96. NeurIPS 2024Segment Anything without Supervision
97. NeurIPS 2024Smoothie: Label Free Language Model Routing
98. NeurIPS 2024Stronger Than You Think: Benchmarking Weak Supervision on Realistic Tasks
99. NeurIPS 2024The ALCHEmist: Automated Labeling 500x CHEaper than LLM Data Annotators
100. NeurIPS 2024The Many Faces of Optimal Weak-to-Strong Learning

FAQ

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

136 in total: 34 in 2023, 27 in 2024, 30 in 2025, 45 in 2026.

How many Weak Supervision papers are at NeurIPS 2026?

NeurIPS 2026 has 12 Weak Supervision papers, versus 15 the year before.

How many Weak Supervision papers are at ICLR 2026?

ICLR 2026 has 8 Weak Supervision papers, versus 9 the year before.

How many Weak Supervision papers are at ICML 2026?

ICML 2026 has 15 Weak Supervision papers, versus 2 the year before.

How many Weak Supervision papers are at CVPR 2026?

CVPR 2026 has 10 Weak Supervision papers, versus 4 the year before.

Which conference publishes the most Weak Supervision papers?

NeurIPS, with 42 papers from 2023 to 2026.