MedElevator Войти в исследования
← Поиск публикаций
Медицинская библиотека · Proceedings of machine learning research · 2026

Contrastive Diffusion Alignment: Learning Structured Latents for Controllable Generation.

Sandilya Ruchi, Perez Sumaira, Lynch Charles, Victoria Lindsay, Zebley Benjamin, Buchanan Derrick Matthew, Bhati Mahendra T, Williams Nolan, Spellman Timothy J, Gunning Faith M, Liston Conor, Grosenick Logan

СтатьяПеревод пока не подготовлен

Аннотация

Diffusion models excel at generation, but their latent spaces are high dimensional and not explicitly organized for interpretation or control. We introduce ConDA (Contrastive Diffusion Alignment), a plug-and-play geometry layer that applies contrastive learning to pretrained diffusion latents using auxiliary variables (e.g., time, stimulation parameters, facial action units). ConDA learns a low-dimensional embedding whose directions align with underlying dynamical factors, consistent with recent contrastive learning results on structured and disentangled representations. In this embedding, simple nonlinear trajectories support smooth interpolation, extrapolation, and counterfactual editing while rendering remains in the original diffusion space. ConDA separates editing and rendering by lifting embedding trajectories back to diffusion latents with a neighborhood-preserving kNN decoder and is robust across inversion solvers. Across fluid dynamics, neural calcium imaging, therapeutic neurostimulation, facial expression dynamics, and monkey motor cortex activity, ConDA yields more interpretable and controllable latent structure than linear traversals and conditioning-based baselines, indicating that diffusion latents encode dynamics-relevant structure that can be exploited by an explicit contrastive geometry layer.

Для цитирования: Sandilya Ruchi, Perez Sumaira, Lynch Charles, Victoria Lindsay, Zebley Benjamin, Buchanan Derrick Matthew, Bhati Mahendra T, Williams Nolan, Spellman Timothy J, Gunning Faith M, Liston Conor, Grosenick Logan. Contrastive Diffusion Alignment: Learning Structured Latents for Controllable Generation.. Proceedings of machine learning research. 2026.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…

История · было → стало

Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.

Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.