Skip to main content
Search

Institut national de recherche en sciences et technologies du numérique

4 related articles

Subscribe to updates
  1. arXiv

    Discrete diffusion need not pay for vocabulary size: masked and uniform noising both hit minimax rates on sparse data

    Discrete diffusion models have lacked the statistical theory that continuous diffusion enjoys, and it was unclear whether they can exploit sparse structure in the data distribution. This analysis casts masked-diffusion score estimation as a per-context label-smoothed classifier and shows the smoothing is equivalent to regularizing with the score loss of the uniform distribution. On sparse distributions where at most k sequences have positive probability, the terminal laws of both masked and uniform noising attain the sparse minimax rate, and the smoothing level can be chosen adaptively from the number of distinct training sequences without knowing the sparsity parameter.

  2. arXiv

    Kernel solvers stop choking on preconditioner memory: sketched blocks push model size to ten million parameters with better accuracy

    The preconditioner memory of kernel ridge regression no longer grows with model size, letting Bloks scale to ten million parameters and match or beat state-of-the-art accuracy on large-scale benchmarks such as Higgs.

  3. ArchesClimate-SSP, trained with an energy-score loss, reaches 0.9739 K land-temperature RMSE on unseen SSP5-3.4, beating MESMER-M's 1.0963 K

    The study introduces ArchesClimate-SSP (AC-SSP), built on the ArchesWeather architecture and trained with an energy-score loss plus a variogram loss while conditioning on forcing concentrations (CO2, CH4, N2O, six aerosol species, ozone), and trained on IPSL-CM6A-LR monthly data for 2015-2100 to autoregressively generate SSP scenarios unseen in training; on the held-out overshoot scenario SSP5-3.4 its 2090-2100 land surface temperature RMSE is 0.9739 K, lower than the statistical emulator MESMER-M's 1.0963 K, and its interannual variability of 0.6050 is closer to the IPSL target's 0.5166 than MESMER-M's 0.6375.

  4. Auditing Sex/Gender Disparities in Emergency Triage with LLM-based Paired Comparisons

    The study introduces a domain-agnostic paired-comparison approach that fine-tunes a large language model to emulate documented emergency triage decisions and then compares predictions on sex-swapped pairs in which only sex is flipped while documented clinical content is held constant, finding that otherwise identical presentations were more likely to receive a lower-severity (less urgent) predicted triage score as female than male, with a pooled per-pair rate of about 1.1% (95% CI 0.9–1.3) across more than 140,000 Bordeaux University Hospital admissions and a directionally consistent but larger 2.2% (1.7–2.7) in MIMIC-IV, while a model retrained on sex-neutralized inputs eliminated the between-sex prediction gap, indicating the asymmetry is mediated by explicit sex markers.

You’re all caught up