Skip to main content
Search

McGill University

14 related articles

Subscribe to updates
  1. arXiv

    Autonomous-driving tests stop following a fixed script: a scenario orchestrator rewrites other vehicles in real time and fully realizes the intended interaction in all six scenario families

    Autonomous-driving policies react in closed loop, so a scenario scripted with fixed routes often drifts away from the interaction it was meant to stage, a problem called closed-loop drift. A scenario orchestrator continuously monitors the joint traffic state and adapts non-ego actors through re-timing, re-routing and role reassignment, solving a constrained optimization at each step so the test intent survives while interventions stay limited and traffic stays feasible. In the CARLA simulator across six scenario families and six ego policies, the method fully realizes the intended interaction in every scenario family, while baselines such as ProSim, LASER and OSC2Runner mostly satisfy only the two easiest scenes.

  2. arXiv

    Risk-controlled candidate screening drops conformal p-values: UNEC calibrates the cutoff directly, taking about 1.3 seconds per repetition over the full simulation grid versus 35 to 37 for SCoRE

    Machine learning screens far more candidates than can be experimentally validated, so selection must control risk before validation. UNEC normalizes predictive margins by local uncertainty and calibrates the selection cutoff directly on held-out losses, without building conformal p- or e-values, and supports general bounded losses and observation-specific thresholds. On solubility, toxicity and lipophilicity screening it holds risk-power trade-offs competitive with conformal selection, the NP method and SCoRE variants; in simulation the full grid took about 1.3 seconds per repetition versus 35 to 37 seconds for SCoRE.

  3. Retrained on Nigerian scans, a brain-age model reaches 9.82 years of error, then 8.78 on a second cohort

    Brain-age models trained mostly on Global North data lose accuracy when applied to Nigerian adults. A population-specific Gaussian process regression model trained on clinical T1-weighted MRI from 309 Nigerian adults reached a mean absolute error of 9.82 years on a 47-participant held-out test set without age-bias correction; on the 44-participant head-to-head set it reached 10.24 years, against 10.52 years for PyBrainAge, 12.33 for SynthBA, 17.51 for ENIGMA-Brain Age and 7.87 for SPARROW. After correction it reached 6.39 years internally but 8.78 years on a separate Nigerian cohort, where every model's R2 was negative.

  4. arXiv

    Multimodal pretraining stops guessing: 268 runs yield a data-mixing law for text, vision and audio

    A study of text, vision and audio mixture-of-experts pretraining fits a data-mixing scaling law from 268 runs and draws actionable conclusions: audio should spend compute on parameters, repeating vision or audio data beyond two passes buys little, and vision interferes most with audio.

  5. Brain maps now decode to tasks, concepts and domains at once, not just a task name

    Functional decoding of brain activation maps has a new route: NiCLIP aligns neuroimaging article text with coordinate-derived activation maps in one shared latent space through contrastive learning, then uses a cognitive ontology to turn similarity into posterior probabilities over tasks, concepts and domains. Widely used correlation decoders and GC-LDA rely on precomputed meta-analytic maps and fixed vocabularies and give task-level predictions only. On Human Connectome Project group-level maps across seven domains, the best configuration reached 62.86% task recall against a 20.71% maximum for Neurosynth and GC-LDA, while the same model dropped markedly on subject-level maps.

  6. bioRxiv

    An old ovary can be rewritten by its host: aged grafts ovulate and produce live pups in young mice

    Ovarian aging is not a one-way, fixed process. Using an ovary-specific transcriptomic clock, OvaAge, alongside epigenetic and pathway methylation clocks, researchers transplanted old ovaries into young mice and young ovaries into old mice and found that ovarian molecular age shifts in both directions with host age, but only partially. Ovaries from 13-month-old donors supported full-term pregnancy and live births at 17-18 months of age in young hosts, whereas ovaries of the same age failed to reproduce in old hosts.

  7. arXiv

    Multi-agent systems homogenize in code generation, hiring, and peer review, and neither sampling stochasticity nor mixed-model configurations mitigates the risk

    The study operationalizes homogenization in multi-agent systems through conformity, polarization, and inertia, evaluates it on code generation, hiring, and peer review, and finds that agent interactions reduce diversity and amplify shared failures, while sampling stochasticity and mixed-model configurations fail to mitigate these risks.

  8. arXiv

    GrayShield overwrites Transformer weight mantissa LSBs with a Gray-code-guided low-transition sequence, cutting attacker recovery by about 50 percentage points across four model presets and two real-world malware payloads while keeping accuracy impact under 1%

    The work proposes GrayShield, a lightweight, post-training, zero-data sanitization method that completely overwrites the declared mantissa least-significant-bit channel of 32-bit floating-point Transformer weights with a Gray-code-guided low-transition sequence, giving that channel zero capacity; across four Transformer model presets, two real-world malware payloads, and five attacker variants, it maintains sub-1% accuracy impact and achieves a 49.96±0.66 percentage-point Recovery Reduction against seven post-training defenses, with substantially smaller weight-distribution shift than PatternMask and Post-Training Quantization.

  9. medRxiv

    Extending the training window from the final hour to the full duration of labor and adding time from labor onset let a model identify 40.3% of hypoxic-ischemic encephalopathy cases at least 3 hours before delivery, an 8.4-point gain over the last-hour classifier

    Using 174,186 deliveries at gestational age ≥35 weeks from Kaiser Permanente Northern California, the study extracted 40 cardiotocography (CTG) features from 20-minute epochs to train Random Forest classifiers targeting severe acidosis (N=2,636) and clinically validated hypoxic-ischemic encephalopathy (HIE, N=304), progressively extended the training window from the final hour before delivery to the full duration of labor, and added time from labor onset (TLO) as a feature; evaluated continuously at a fixed 15% false positive rate, early identification improved as the window was extended and plateaued at 18 hours, and adding TLO gave the best early-warning performance, with 40.3% of HIE cases identified at least 3 hours and 27.3% at least 6 hours before delivery, absolute gains of 8.

  10. A 45-member modified Delphi consensus: screen for eating disorders before GLP-1 receptor agonist prescribing, and generally avoid these agents in active anorexia and bulimia nervosa

    Using a three-round modified Delphi method, a 45-member panel of clinicians and researchers in diabetes, obesity and/or eating disorders plus people with lived experience reached an average 91.7% consensus on recommendations for GLP-1 receptor agonist use in eating disorder contexts, covering pre-treatment screening, risk-stratified monitoring, standardized educational materials and concurrent evidence-based psychotherapy, stating that these agents should generally be avoided in active eating disorders, particularly anorexia nervosa, atypical anorexia nervosa and bulimia nervosa with marked dietary restraint and/or significant weight suppression, and setting out research priorities in epidemiology, regulation and binge eating disorder trials.