mlsanityData to 7 Oct 2026

Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models

arXiv 2402.03749 ↗First posted 6 Feb 2024cs.CVcited by 28

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

Recent advancements in large language models have sparked interest in their extraordinary and near-superhuman capabilities, leading researchers to explore methods for evaluating and optimizing these abilities, which is called superalignment. In this context, our paper delves into the realm of vision foundation models, focusing on the concept of weak-to-strong generalization, which involves using a weaker model to supervise a stronger one, aiming to enhance the latter's capabilities beyond the former's limits. We introduce a novel and adaptively adjustable loss function for weak-to-strong supervision. Our comprehensive experiments span various scenarios, including few-shot learning, transfer learning, noisy label learning, and common knowledge distillation settings. The results are striking: our approach not only exceeds the performance benchmarks set by strong-to-strong generalization but also surpasses the outcomes of fine-tuning strong models with whole datasets. This compelling evidence underscores the significant potential of weak-to-strong generalization, showcasing its capability to substantially elevate the performance of vision foundation models. The code is available at https://github.com/ggjy/vision_weak_to_strong.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 33 of its 41 references that are in the map. First posted on arXiv 6 Feb 2024.

Builds on

33 of 41 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision14 Dec

Other references · 32

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020Language Models are Few-Shot Learnerscited by 32,199
  5. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  6. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  7. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  8. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  9. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  10. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  11. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  12. 2023Segment Anythingcited by 8,158
20 more
  1. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  2. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  3. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  4. 2016Wide Residual Networkscited by 3,057
  5. 2016Matching Networks for One Shot Learningcited by 2,246
  6. 2014FitNets: Hints for Thin Deep Netscited by 1,619
  7. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  8. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,194
  9. 2019Relational Knowledge Distillationcited by 873
  10. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 691
  11. 2018Born Again Neural Networkscited by 539
  12. 2019A Comprehensive Overhaul of Feature Distillationcited by 341
  13. 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
  14. 2021Distilling Knowledge via Knowledge Reviewcited by 285
  15. 2023Sequential Modeling Enables Scalable Learning for Large Vision Modelscited by 212
  16. 2020Meta-Baseline: Exploring Simple Meta-Learning for Few-Shot Learningcited by 179
  17. 2022A Survey of Vision-Language Pre-Trained Modelscited by 177
  18. 2022Selective-Supervised Contrastive Learning with Noisy Labelscited by 112
  19. 2023One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillationcited by 77
  20. 2023VanillaKD: Revisit the Power of Vanilla Knowledge Distillation from Small Scale to Large Scalecited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 11; 2025: 14; 2026: 3.

28 papers.

Sort:

By area:

  1. 2025Muddit: Liberating Generation Beyond Text-to-Image with a Unified Discrete Diffusion Modelcited by 52
  2. 2024What is the Role of Small Models in the LLM Era: A Surveycited by 41
  3. 2024How Far Are We From AGI: Are LLMs All We Need?cited by 31
  4. 2024Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalizationcited by 23
  5. 2024MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimizationcited by 20
  6. 2025Scaling Laws For Scalable Oversightcited by 18
  7. 2024Provable Weak-to-Strong Generalization via Benign Overfittingcited by 16
  8. 2024The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignmentcited by 11
  9. 2024Hummer: Towards Limited Competitive Preference Datasetcited by 10
  10. 2025Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimensioncited by 10
  11. 2024Defending LVLMs Against Vision Attacks through Partial-Perception Supervisioncited by 7
  12. 2025On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspectivecited by 7
16 more
  1. 2024Bayesian WeakS-to-Strong from Text Classification to Generationcited by 6
  2. 2025Standardizing Intelligence: Aligning Generative AI for Regulatory and Operational Compliancecited by 5
  3. 2025Generalizable Video Quality Assessment via Weak-to-Strong Learningcited by 5
  4. 2024Weak-to-Strong Generalization beyond Accuracy: a Pilot Study in Safety, Toxicity, and Legal Reasoningcited by 4
  5. 2025EnsemW2S: Enhancing Weak-to-Strong Generalization with Large Language Model Ensemblescited by 4
  6. 2025Weak-to-Strong Generalization under Distribution Shiftscited by 4
  7. 2025Weak-to-Strong Generalization via Bregman Bias-Variance Decompositioncited by 3
  8. 2025The Capabilities and Limitations of Weak-to-Strong Generalization: Generalization and Calibrationcited by 2
  9. 2024EnsemW2S: Enhancing Weak-to-Strong Generalization with Large Language Model Ensemblescited by 1
  10. 2025Alice: Proactive Learning with Teacher's Demonstrations for Weak-to-Strong Generalizationcited by 1
  11. 2025Does Weak-to-strong Generalization Happen under Spurious Correlations?cited by 1
  12. 2025Denoising Mutual Knowledge Distillation in Bi-Directional Multiple Instance Learningcited by 0
  13. 2025Exploring Weak-to-Strong Generalization for CLIP-based Classificationcited by 0
  14. 2026Weak-to-Strong Knowledge Distillation Accelerates Visual Learningcited by 0
  15. 2026On the Blessing of Pre-training in Weak-to-Strong Generalizationcited by 0
  16. 2026Gradient Transformer: Learning to Generate Updates for LLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Combining inherent knowledge of vision-language models with unsupervised domain adaptation through strong-weak guidance
  2. 2023Robust Fine-Tuning of Vision-Language Models for Domain Generalization
  3. 2024A transfer learning framework for weak-to-strong generalization
  4. 2024Improving Weak-to-Strong Generalization with Reliability-Aware Alignment
  5. 2026Dive Into the Implicit Biases of Low-rank Vision-language Alignment
  6. 2025TinyAlign: Boosting Lightweight Vision-Language Models by Mitigating Modal Alignment Bottlenecks
  7. 2025Selective Weak-to-Strong Generalization
  8. 2024Anchor-based Robust Finetuning of Vision-Language Models