Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions

• URL: http://arxiv.org/abs/2502.00620v2
• Date: Wed, 05 Feb 2025 00:36:00 GMT
• Title: Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions
• Authors: Yihao Xue, Jiping Li, Baharan Mirzasoleiman,
• Abstract summary: Weak-to-Strong Generalization (W2SG) serves as an important analogy for understanding how humans might guide superhuman intelligence in the future. We show that W2SG can be characterized using kernels derived from the principal components of weak and strong models' internal representations.
• Score: 12.956498486569103
• License:
• Abstract: Weak-to-Strong Generalization (W2SG), where a weak model supervises a stronger one, serves as an important analogy for understanding how humans might guide superhuman intelligence in the future. Promising empirical results revealed that a strong model can surpass its weak supervisor. While recent work has offered theoretical insights into this phenomenon, a clear understanding of the interactions between weak and strong models that drive W2SG remains elusive. We investigate W2SG through a theoretical lens and show that it can be characterized using kernels derived from the principal components of weak and strong models' internal representations. These kernels can be used to define a space that, at a high level, captures what the weak model is unable to learn but is learnable by the strong model. The projection of labels onto this space quantifies how much the strong model falls short of its full potential due to weak supervision. This characterization also provides insights into how certain errors in weak supervision can be corrected by the strong model, regardless of overfitting. Our theory has significant practical implications, providing a representation-based metric that predicts W2SG performance trends without requiring labels, as shown in experiments on molecular predictions with transformers and 5 NLP tasks involving 52 LLMs.

Related papers

• Understanding the Capabilities and Limitations of Weak-to-Strong Generalization [40.793180521446466]
  We provide theoretical insights into weak-to-strong generalization. We show that the weak model should demonstrate strong generalization performance and maintain well-calibrated predictions. We extend the work of Charikar et al. (2024) to a loss function based on Kullback-Leibler divergence.
  arXiv  Detail & Related papers  (2025-02-03T15:48:28Z)
• Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss [4.4505368723466585]
  We study weak-to-strong generalization for convex combinations of $k$ strong models in the strong class. We obtain a similar misfit-based characterization of performance gain, upto an additional error term that vanishes as $k$ gets large.
  arXiv  Detail & Related papers  (2025-01-31T12:57:58Z)
• Debate Helps Weak-to-Strong Generalization [68.70065254564642]
  We investigate ways of improving human supervision with a strong pretrained model and then supervise the strong model with enhanced weak human supervision. We find that debate can assist a weak model in extracting trustworthy information from an untrustworthy strong model. Experiments on the OpenAI weak-to-strong NLP benchmarks show that the combination approach leads to better alignment.
  arXiv  Detail & Related papers  (2025-01-21T05:36:13Z)
• Super(ficial)-alignment: Strong Models May Deceive Weak Models in Weak-to-Strong Generalization [68.62228569439478]
  We investigate whether there exists an issue of weak-to-strong deception. We find that the deception intensifies as the capability gap between weak and strong models increases. Our work highlights the urgent need to pay more attention to the true reliability of superalignment.
  arXiv  Detail & Related papers  (2024-06-17T11:36:39Z)
• Quantifying the Gain in Weak-to-Strong Generalization [14.453654853392619]
  We show that the improvement in performance achieved by strong models over their weaker counterparts is quantified by the misfit error incurred by the strong model on labels generated by the weaker model. For instance, we can predict the amount by which the strong model will improve over the weak model, and also choose among different weak models to train the strong model, based on its misfit error.
  arXiv  Detail & Related papers  (2024-05-24T00:14:16Z)
• Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models [55.919653720979824]
  This paper focuses on the concept of weak-to-strong generalization, which involves using a weaker model to supervise a stronger one. We introduce a novel and adaptively adjustable loss function for weak-to-strong supervision. 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.
  arXiv  Detail & Related papers  (2024-02-06T06:30:34Z)
• Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision [55.196139002977525]
  Superhuman models will behave in complex ways too difficult for humans to reliably evaluate. Can weak model supervision elicit the full capabilities of a much stronger model? We find that when we naively finetune strong pretrained models on labels generated by a weak model, they consistently perform better than their weak supervisors.
  arXiv  Detail & Related papers  (2023-12-14T23:07:33Z)
• Does Deep Learning Learn to Abstract? A Systematic Probing Framework [69.2366890742283]
  Abstraction is a desirable capability for deep learning models, which means to induce abstract concepts from concrete instances and flexibly apply them beyond the learning context. We introduce a systematic probing framework to explore the abstraction capability of deep learning models from a transferability perspective.
  arXiv  Detail & Related papers  (2023-02-23T12:50:02Z)
• Clustering Effect of (Linearized) Adversarial Robust Models [60.25668525218051]
  We propose a novel understanding of adversarial robustness and apply it on more tasks including domain adaption and robustness boosting. Experimental evaluations demonstrate the rationality and superiority of our proposed clustering strategy.
  arXiv  Detail & Related papers  (2021-11-25T05:51:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.