Skip to main content
Search

University of Pittsburgh

6 related articles

Subscribe to updates
  1. arXiv

    Multi-agent systems stop sharing one workflow: OOPMAS builds agents and workflows per query, reaching 89.6% on mixed tasks

    Automated design of multi-agent systems has mostly worked at the task level, optimizing one fixed workflow per benchmark and applying it to every query. OOPMAS moves design to the individual query: a frozen LLM as meta-agent first generates object-style agents with roles, tools and persistent state, then an executable main-function workflow, while a dynamic skill library accumulates structured lessons from execution feedback without gradient updates. On a benchmark mixing code, math and QA, OOPMAS reaches 89.6% overall accuracy, 18.1 percentage points above the strongest baseline, ADAS.

  2. arXiv

    SAM 3 at 4-bit weights stops grabbing the wrong object: SPQ lifts RefCOCO hit rate from 42.1 to 72.5

    Quantizing SAM 3's image-encoder weights to 4 bits while keeping activations at 8 bits produces prompt-alignment drift: masks look plausible but land on the wrong object or lose the target across video frames. SPQ gives more calibration weight to parameters that affect text-concept scores, constrains feature error along low-variance residual directions, and folds a mean-error correction estimated from paired video rollouts into existing biases. Under W4A8, RefCOCO Pr@0.5 rises from 42.1 for PTQ4SAM to 72.5 while retaining 54.8 COCO mask AP, with no added inference operators.

  3. arXiv

    Neurons have no positions, yet they can be coarse-grained: a renormalization-group map predicts five grokking exponents within 1%–10% of direct fits

    Deep networks on modular addition memorize the training set and generalize only long afterward, a delay called grokking. Treating training epoch as temperature and weight decay as a conjugate field, the study builds neuron-to-neuron couplings from second derivatives of the loss in per-neuron norm coordinates, orders neurons along the Fiedler vector of that coupling graph, and runs a real-space renormalization group on the resulting one-dimensional chain. Five critical exponents from the blocking map agree with values fitted directly from 30 independent training runs to within 1%–10%.

  4. bioRxiv

    A single brief dynamic amplitude-modulated envelope-following response plus machine learning reads out cochlear neural degeneration in gerbils and transfers to human listeners

    The authors tested a dynamic amplitude-modulated (dAM) envelope-following response (EFR) that sweeps the full modulation spectrum in a single brief stimulus, found selective deficits at fast modulation rates without threshold elevation in Mongolian gerbils with histologically verified cochlear neural degeneration (CND), trained a machine-learning classifier that distinguished young from middle-aged animals with high accuracy and whose most informative feature (power near 400-500 Hz) tracked synapse counts, and applied the gerbil-trained classifier without retraining to 56 human listeners, where it separated age groups above chance.

  5. Nature

    A Six-Million-Cell Map of Gene Activity in the Human Prefrontal Cortex: How the PsychAD Cohort and Dreamlet Push Brain-Disease Research to Population Scale

    A single-nucleus RNA-sequencing atlas of the human dorsolateral prefrontal cortex built from nearly 1,500 donors and more than 6.3 million nuclei, together with the companion statistical tool Dreamlet, characterizes cell-type-specific transcriptional changes across eight brain disorders, reports shared cross-disease signatures, cell-composition shifts along Alzheimer's disease progression, and marked up-regulation of PTPRG in microglia.

  6. medRxiv

    Task-Specific Quality Gating for Retinal OCT B-Scans: Learned Representations Over Scalar Metrics in Choroid Segmentation

    Using choroid segmentation as a prototype task on 6,076 OCT B-scans from 80 subjects, this study systematically compared scalar no-reference image quality metrics (BRISQUE, NIQE, PIQE, SNR, PSNR), general-purpose ImageNet-pretrained representations, and retinal foundation models as task-specific quality gates, finding that scalar metrics correlate weakly with segmentation Dice (|r| < 0.20), that general-purpose pretrained representations reach linear-probe ROC-AUC up to about 0.77, that the OCT-specific foundation model RETFound reaches about 0.81, and that only the retinal foundation model embeddings form quality-aligned unsupervised K-Means clusters exceeding a patient-level permutation null.

You’re all caught up