This work identifies which SSL feature families lead to best generation performance and shows that matching stability is a quantitative criterion for selecting them, and shows that using Inception as the feature extractor can improve FID while degrading matching stability and sample quality, revealing a form of metric...
Hugues Van Assel, E. De Brouwer, Saeed Saremi et al.· arXiv.org· 1 citation
As agents rapidly evolve, existing benchmarks can become saturated, limiting their ability to distinguish capabilities and reveal remaining failure modes. Particularly in scientific domains, constructing and updating benchmarks requires substantial time, labor, and domain expertise, making it difficult to keep evaluati...
Dongki Kim, Namkyeong Lee, Surag Nair et al.· 0 citations
Riemannian Wasserstein Entropic Flow Matching (RWEFM) is introduced, a generative framework on the Wasserstein space of a Riemannian manifold that can generate whole single-cell samples in hyperspherical latent spaces and protein conformational ensembles on the torus by respecting the intrinsic geometry of the data.
D. Haviv, E. De Brouwer, Rishabh Anand et al.· 0 citations
AssayLoop is introduced, a sequential experimental design framework combining AssayFormer, a transformer-based amortized acquisition policy trained across historical screens to adapt from experimental feedback, with LLM-derived biological priors through an adaptive handoff to demonstrate the value of learning acquisiti...
Carl N. Edwards, E. De Brouwer, Xin-Er Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.