探索式建模：解鎖第三預訓練軸與端到端生成

摘要

深度學習革命由 AlexNet 開啟，教會我們端到端訓練優於將問題分解為手工設計的階段。然而，生成建模一直是個例外——儘管生成模型能力顯著，它們仍未被端到端訓練。這是因為生成建模的核心在於處理具有多個模態的分佈，而現有的可擴展方法皆以相同方式處理此問題：將生成過程分解，這阻礙了端到端生成。在本工作中，我們提出探索式建模（Explorative Modeling），一種新典範，其分解的是訓練迴圈而非生成過程：探索模型生成與資料之間的 K 個候選匹配，並在最佳匹配上訓練，使預測專注於模態而非模糊化它們。我們發現探索式模型（XMs）在兩種情境下具有價值。首先，增加探索為現有生成模型增添了一條超越參數與資料的第三條預訓練軸線——在連續與離散領域（影像、視訊與語言）中，擴展探索能單調地提升效能。值得注意的是，探索帶來的增益隨規模擴大而增加：隨資料擴展從 7% 攀升至 36%，隨模型成長從 13% 提升至 23%，而在 3 倍計算量下的效率增益更超過兩倍。具體而言，探索使 FLOP 效率提升 4.1 倍、樣本效率提升 6.2 倍、參數效率提升 47%，將最強的影像生成方案推升至 ImageNet 上無引導的近最佳 1.43 FID，實現現有模型端到端程度的擴展，並解鎖擴展泛化能力。其次，XM 實現了端到端的重建式生成建模，在控制任務上以 16–256 倍更少的推論步驟媲美擴散模型。綜合而言，這些結果確立了 XM 既作為現有生成模型的新預訓練軸線，也作為一種獨立的端到端生成建模典範。

English

The deep learning revolution, kicked off by AlexNet, taught us that end-to-end training beats decomposing a problem into hand-designed stages. Generative modeling, however, has remained the exception-despite generative models being remarkably capable, they are still not trained end-to-end. This is because, at its core, generative modeling is about handling distributions with many modes, and existing scalable approaches handle this the same way, by factoring the generation procedure, which prevents end-to-end generation. In this work, we introduce Explorative Modeling, a new paradigm that instead factors the training loop, exploring K candidate matches between model generations and data, and training on the best, so predictions commit to modes rather than blurring them. We find Explorative Models (XMs) useful in two settings. First, increasing exploration adds a third pretraining axis beyond parameters and data for existing generative models-where scaling exploration monotonically improves performance across both continuous and discrete domains (images, video, and language). Notably, gains from exploration increase with scale, climbing from 7% to 36% as data scales and from 13% to 23% as models grow, with efficiency gains more than doubling at 3x the compute. Concretely, exploration improves FLOP efficiency by 4.1x, sample efficiency by 6.2x, parameter efficiency by 47%, lifts the strongest of image-generation recipes to a near-state-of-the-art 1.43 FID on ImageNet without guidance, enables scaling how end-to-end existing models are, and unlocks scaling generalization. Second, XMs enable end-to-end reconstructive generative modeling, matching diffusion on control tasks with 16-256x fewer inference steps. Together, these results establish XMs as both a new pretraining axis for existing generative models and a standalone end-to-end generative modeling paradigm.