Russian Academy of Sciences

4 related articles

Subscribe to updates

researchsquare.com Nov 9, 2026
Evidential deep learning hits an evidence ceiling: as confidence nears one, the active set collapses to a single class
Evidential deep learning represents classification uncertainty through Dirichlet parameters, but the standard objective shows evidence contraction at high confidence. This work analyzes the optimization landscape of the regularized Dirichlet likelihood on a box domain, proves that the second-order sufficient conditions reduce to a scalar self-consistency derivative being positive, and gives an enumeration algorithm with a finite-convergence guarantee. On logits from a ResNet-18 trained on CIFAR-10, total evidence rises monotonically with confidence yet never exceeds a theoretical threshold near 19.75, while the active set shrinks from about 2.05 classes to one.
Evidential deep learning represents classification uncertainty through Dirichlet parameters, but the standard objective shows evidence contraction at high confidence. This work analyzes the optimization landscape of the regularized Dirichlet likelihood on a box domain, proves that the second-order sufficient conditions reduce to a scalar self-consistency derivative being positive, and gives an enumeration algorithm with a finite-convergence guarantee. On logits from a ResNet-18 trained on CIFAR-10, total evidence rises monotonically with confidence yet never exceeds a theoretical threshold near 19.75, while the active set shrinks from about 2.05 classes to one.
Saratov State University Russian Academy of Sciences
Danila Ozerov（First author, Corresponding author）
Evidential Deep Learning Constrained Optimization Uncertainty Quantification Hybrid Compact-Core Newton Evidence Ceiling
International Journal of Corrosion and Scale Inhibition Sep 28, 2026
Andreev and Vesely, in dialogue with ChatGPT, propose a refined definition of "corrosion inhibitor" that covers the protection after-effect of pretreatment and separates inhibition from conversion treatment and bulk-phase coatings
In a question-and-answer dialogue with ChatGPT, the authors compare normative definitions from GOST 9.106–2021, NACE/ASTM G193-22 and ISO 8044:2024, point to the ISO requirement that an inhibitor be "present in the corrosion system" and to the absence of a distinction between the bulk environment and the near-surface region, and then define an inhibitor through its action on the metal or the corrosion system, offering working definitions of a chemisorbed layer, a conversion coating and a bulk-phase anticorrosion coating, and arriving at a refined definition that excludes substances whose action significantly changes the concentration of corrosive components in the bulk environment, forms a conversion coating, or involves applying a bulk-phase anticorrosion coating.
In a question-and-answer dialogue with ChatGPT, the authors compare normative definitions from GOST 9.106–2021, NACE/ASTM G193-22 and ISO 8044:2024, point to the ISO requirement that an inhibitor be "present in the corrosion system" and to the absence of a distinction between the bulk environment and the near-surface region, and then define an inhibitor through its action on the metal or the corrosion system, offering working definitions of a chemisorbed layer, a conversion coating and a bulk-phase anticorrosion coating, and arriving at a refined definition that excludes substances whose action significantly changes the concentration of corrosive components in the bulk environment, forms a conversion coating, or involves applying a bulk-phase anticorrosion coating.
Russian Academy of Sciences
Corrosion inhibitor definition ChatGPT Conversion coating
Frontiers in Cellular and Infection Microbiology Sep 28, 2026
Knocking out M. tuberculosis small RNAs MTS0997 and MTS1338 shortens mouse survival without changing lung bacterial burden
The authors built unmarked single- and double-knockout Mycobacterium tuberculosis strains lacking MTS0997, MTS1338, or both, profiled whole-genome expression during culture growth, in IFN-γ-activated and non-activated murine bone marrow-derived macrophages, and in infection of highly TB-susceptible I/St mice, and found that MTS0997 deletion had the larger impact on bacterial biology (altered succinate/fumarate respiration, up-regulated ESX-1 secretion genes, increased virulence in mice, reduced lung pro-inflammatory cytokine production), that all three knockout strains killed mice significantly faster than wild type while lung CFU counts at 6 weeks did not differ, and that B-cell and MHC-II-expressing lung cell populations differed between single- and double-knockout infections, suggesting
The authors built unmarked single- and double-knockout Mycobacterium tuberculosis strains lacking MTS0997, MTS1338, or both, profiled whole-genome expression during culture growth, in IFN-γ-activated and non-activated murine bone marrow-derived macrophages, and in infection of highly TB-susceptible I/St mice, and found that MTS0997 deletion had the larger impact on bacterial biology (altered succinate/fumarate respiration, up-regulated ESX-1 secretion genes, increased virulence in mice, reduced lung pro-inflammatory cytokine production), that all three knockout strains killed mice significantly faster than wild type while lung CFU counts at 6 weeks did not differ, and that B-cell and MHC-II-expressing lung cell populations differed between single- and double-knockout infections, suggesting
Russian Academy of Sciences
Elena G. Salina（First author）
MTS0997 MTS1338 RNA Sequencing Host immune responses Virulence
Journal of Chemical Information and Modeling Sep 14, 2026
In-Context Learning Meets Small Molecule Property Prediction: Benchmarking Novel Machine Learning Approaches
This study comprehensively benchmarked tabular foundation models (TFMs) based on in-context learning for small organic molecule property prediction, comparing several TFMs with multiple machine learning methods across 11 data sets (regression, random and structure-aware splits, up to 10,000 molecules each), and found that TFMs consistently outperform XGBoost, CatBoost, multilayer perceptrons, and other descriptor-based methods even with careful hyperparameter selection for the baselines, achieve accuracy on par with or better than graph-based methods including those pretrained on chemical data, with Uni-Mol2 slightly outperforming TFMs in some experiments, while retrieval (selecting the 500 closest neighbors by Tanimoto similarity for each test molecule) yields further improvement on relat
This study comprehensively benchmarked tabular foundation models (TFMs) based on in-context learning for small organic molecule property prediction, comparing several TFMs with multiple machine learning methods across 11 data sets (regression, random and structure-aware splits, up to 10,000 molecules each), and found that TFMs consistently outperform XGBoost, CatBoost, multilayer perceptrons, and other descriptor-based methods even with careful hyperparameter selection for the baselines, achieve accuracy on par with or better than graph-based methods including those pretrained on chemical data, with Uni-Mol2 slightly outperforming TFMs in some experiments, while retrieval (selecting the 500 closest neighbors by Tanimoto similarity for each test molecule) yields further improvement on relat
Russian Academy of Sciences
Dmitriy D. Matyushin（First author）
Tabular Foundation Models TabPFN In-Context Learning Molecular Property Prediction Retrieval-Augmented Inference

You’re all caught up

Evidential deep learning hits an evidence ceiling: as confidence nears one, the active set collapses to a single class

Andreev and Vesely, in dialogue with ChatGPT, propose a refined definition of "corrosion inhibitor" that covers the protection after-effect of pretreatment and separates inhibition from conversion treatment and bulk-phase coatings

Knocking out M. tuberculosis small RNAs MTS0997 and MTS1338 shortens mouse survival without changing lung bacterial burden

In-Context Learning Meets Small Molecule Property Prediction: Benchmarking Novel Machine Learning Approaches