Dexterity from Supervised Play

New research challenges the superposition theory of polysemanticity, suggesting neural networks are more interpretable than previously believed through recursive generalization patterns.

Researchers introduce FDE-Gym, a benchmark simulating real-world AI delivery tasks where agents must clarify ambiguous requirements and discover unstated constraints in customer environments.

Researchers introduce T2-z, a method for test-time optimization of environment latents in robotic world models that improves prediction accuracy and action success rates across simulations and real-world tasks.

Researchers introduce OmniVerifier-C1, a decomposed graded verifier that provides continuous scoring instead of binary judgments for multimodal verification tasks.

Researchers propose MirrorMoE to address the exploration dilemma in modern Mixture-of-Experts architectures between fine-grained segmentation and shared experts.