Generative modeling of intrinsically disordered protein regions by reinforcing sparse autoencoder featuresResearchOct 2, 2026
Constitutional Value Potentials: reading and steering internal priority margins in language modelsResearchOct 2, 2026
Compressing History into Memory: Distilling Transformers into Recurrent TransformersResearchOct 2, 2026
Trust the Direction, Search the Step: Zero-and-First-Order Methods for LLM Fine-TuningResearchOct 2, 2026
Cost-augmented Schr\"odinger bridges on graphs are exactly solvable: a Feynman-Kac tilt replaces learned controlResearchOct 2, 2026
From Gradients to Capabilities: Understanding Multi-Teacher On-Policy DistillationResearchOct 2, 2026