AI 对齐世界，靠的不是堆参数，是挑对视角
大模型之间为什么能对齐？主流解释是「规模够大就行」;这篇把猜想变成了定理，而且结论反着来：让模型对齐世界的，不是参数规模，而是你喂给它的「视角」——同一场景的两个好视角，比预测下一个 token 高效 42 倍。研究者证明了「联合嵌入预测架构」只要用同一场景的两个观察当训练对，不同模态的模型就会收敛到同一个世界表征，只差一个旋转；而语言模型之所以要和视觉对齐得靠堆规模，是因为它用的「视角」是文档片段，是坏视角。他们用 940M 张网页图片验证了这套理论，还发现一个 1 亿参数的小文本模型，对齐视觉的能力超过了 720 亿参数的语言模型。这不是你明天能用上的东西，但它改写了「大力出奇迹」的底层叙事：对齐的杠杆在数据视角的设计，不在算力。
📄 原文摘要(英文)
Whether models trained on different modalities converge to the same representation of the world is central to AI. The Platonic Representation Hypothesis (PRH) conjectures that they do as they scale, but without a proof it offers neither a guarantee nor a recipe. We turn the hypothesis into a theorem. First, we prove a Platonic Representation Theorem: Joint-Embedding Predictive Architectures (JEPAs) trained separately on any number of modalities learn the same representation up to one rotation whenever their positive pairs are two observations of one scene--good views--and their misalignment is bounded otherwise. Second, we show that next-token prediction is a JEPA with bad views, parts of a single document, which explains why language models align with vision only through scale. Third, we validate the theory from 145k captions to the texts of 940M web images, and uncover a scaling law of alignment that predicts unseen runs within seed noise. Good views need 42 times fewer training samples than next-token prediction on the same corpus to reach the same alignment, and a 109M-parameter text JEPA aligns with DINOv2 and CLIP's image encoders better than every language model we evaluated, up to 72B parameters, and than CLIP's text encoder: next-token prediction is a remarkably inefficient route to platonic representations. As a showcase, one rotation turns separately trained encoders into an image captioner (CIDEr .838 on COCO). Even that rotation needs no pair: on the latest pair-free benchmark, our text JEPA beats every zero-shot accuracy reported with language models up to 8B parameters, e.g., 64\% against at most 56\% top-5 on ImageNet-100. Positive views, and not scale alone, are thus the lever to build platonic representations.