Tag
This paper introduces Graph-PRefLexOR, a family of graph-native reasoning models fine-tuned with group relative policy optimization (GRPO) to generate traceable scientific hypotheses through explicit reasoning phases. The method achieves 40-65% improvements over base models in reasoning traceability and demonstrates enhanced semantic diversity and conceptual recombination.
DeepRHP is a hybrid variational autoencoder that guides the design of random heteropolymers as protein mimics, demonstrated by stabilizing membrane proteins like Aquaporin Z in non-native environments.
This paper presents a range-aware Bayesian optimization framework that directly scores the posterior probability that a candidate satisfies a target property range, enabling discovery of diverse valid designs across multiple specifications.