arXiv:2605.04356v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards has been used to elicit impressive performance from language models in many domains. But, broadly beneficial deployments of AI may require us to train models with strong capabilities in “fuzzy”, hard-to-supervise domains. In this paper, we develop methods to align language models in fuzzy domains […]
LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stunts
arXiv:2605.05110v1 Announce Type: cross Abstract: Designing reward functions for agile robotic maneuvers in reinforcement learning remains difficult, and demonstration-based approaches often require reference motions that are unavailable for novel platforms or extreme stunts. We present LineRides, a line-guided learning framework that enables a custom bicycle robot to acquire diverse, commandable stunt behaviors from a user-provided […]
Probe-Geometry Alignment: Erasing the Cross-Sequence Memorization Signature Below Chance
arXiv:2605.01699v3 Announce Type: replace-cross Abstract: Recent attacks show that behavioural unlearning of large language models leaves internal traces recoverable by adversarial probes. We characterise where this retention lives and show it can be surgically removed without measurable capability cost. Our central protocol is a leave-one-out cross-sequence probe that tests whether a memorisation signature generalises across […]
Intermediate stages in the origin of metabolism at a phosphorylating hydrothermal vent
arXiv:2510.08410v3 Announce Type: replace Abstract: The origin of life required the emergence of metabolism, an autocatalytic network of enzymatic reactions that synthesize amino acids, nucleotides and cofactors. At the origin of metabolism there were no enzymes–how did it start? Empirical studies addressing early metabolic evolution are lacking. Harnessing protein structures for metabolic enzymes, we identify […]
Intermediate stages in the origin of metabolism at a phosphorylating hydrothermal vent
arXiv:2510.08410v3 Announce Type: replace Abstract: The origin of life required the emergence of metabolism, an autocatalytic network of enzymatic reactions that synthesize amino acids, nucleotides and cofactors. At the origin of metabolism there were no enzymes–how did it start? Empirical studies addressing early metabolic evolution are lacking. Harnessing protein structures for metabolic enzymes, we identify […]
Noise-accelerated Kramers Escape and Coherence Resonance in a 5D Neural Manifold
arXiv:2605.04088v2 Announce Type: new Abstract: Intrinsic channel noise is fundamental to neural processing, yet its state-dependent nature, when constrained by strict Feller boundary conditions, is often overlooked. Here, we demonstrate that this bounded multiplicative noise is not merely a source of jitter but an active dynamical force that fundamentally reshapes neural excitability. Investigating a 5D […]
Insulet CEO details ‘Scrubs’ device promotion win as it boosts awareness, education drive
Insulet has always been a strong leader in the medtech industry when it comes to its marketing game—and the diabetes device maker isn’t looking to rest on its laurels.
End-of-surgery prediction of postoperative infectious complications from intraoperative vital-sign dynamics
npj Digital Medicine, Published online: 07 May 2026; doi:10.1038/s41746-026-02707-1 End-of-surgery prediction of postoperative infectious complications from intraoperative vital-sign dynamics
Hubris in the age of intelligent medicine—from clinical hierarchies to algorithmic amplification
npj Digital Medicine, Published online: 07 May 2026; doi:10.1038/s41746-026-02717-z Hubris in the age of intelligent medicine—from clinical hierarchies to algorithmic amplification
Financial incentives increase uptake and perceived effectiveness of autonomous medical AI, yet patients still seek human reconfirmation
npj Digital Medicine, Published online: 07 May 2026; doi:10.1038/s41746-026-02635-0 Financial incentives increase uptake and perceived effectiveness of autonomous medical AI, yet patients still seek human reconfirmation
Off by a beat: the effects of temporal misalignment in reinforcement learning for sepsis treatment
npj Digital Medicine, Published online: 07 May 2026; doi:10.1038/s41746-026-02625-2 Reinforcement learning shows promise for guiding sequential clinical decisions, yet common data preprocessing introduces temporal misalignment that violates causal assumptions. Using sepsis management as a case study, we demonstrate that such misalignment produces inappropriate treatment recommendations in nearly half of patient states. This widespread methodological flaw […]