archive
Every paper Pith has read. Search by title, abstract, or pith.
548 papers in q-bio.QM · page 2
-
Cold-start selector picks first 3D medical volumes without labels
Dataset-Aware Cold-Start Active Learning for Annotation-Efficient 3D Medical Image Segmentation
-
R package automates validation for team data entry
CollaboratoR: A scalable workflow for collaborative data entry and management
-
ODEs give likelihood for SIR on contact trees from tracing data
Multi-type branching inference on contact trees with application to COVID-19
-
Delays in inhibitory networks create slow rhythms
Dissecting emerging slow rhythms in delay-coupled neural oscillators
-
Biomedical QA scores rise from 65.9 to 71.0 with selective evidence escalation
BioHarness: Substrate-Aware Evidence Assembly for Biomedical Question Answering across Literature, Knowledge Bases, and Biological Atlases
-
Logit alignment conditions bio LMs on contexts
Contextualizing Biological Language Models across Modalities via Logit-Space Contrastive Alignment
-
Decoder turns single-neuron spikes into scene descriptions
Can neurons speak? Semantic narration of vision at single-cell resolution
-
Multimodal data improves glucose forecasts only for complex models
MetaboNet-Bench: A Multi-modal Benchmark for Glucose Forecasting in Type 1 Diabetes
-
Bandit weighting cuts epidemic forecast WIS scores
Adaptive COVID-19 Trajectory Forecasting Using MAB-Inspired Ensemble Weighting
-
Fiducial markers let microfluidic imaging run in real time
DART: A design-aware microfluidic chip paradigm for real-time live-cell image analysis
-
Noise erases nonlinear edge over linear models in biomedicine
Measurement noise limits the advantage of nonlinear models over linear models in biomedical prediction
-
Separates clean molecular properties from noisy labels
MOLAR: Learning Multimodal Molecular Representations from Noisy Labels
-
Multiscale backbone model recovers protein structures from noisy cryo-EM
Multiscale reconstruction of protein conformations from cryo-EM images
-
Transformer directly forecasts multi-step molecular dynamics
ASTEROID: A Spatiotemporal Information Transformer for Forecasting Multi-Step Time Series of Molecular Dynamics
-
Reweighted bridge matching aligns tau PET across sites
Feynman Kac Reweighted Schr\"odinger Bridge Matching for Surface-Based Tau PET Harmonization
-
Unsupervised archetypes flag high N2O emissions in sludge
Archetypal Microbiome Profiles as Indicators of Nitrous Oxide Emission States in Activated Sludge
-
Dual-discriminator GAN generates lab-validated modified antimicrobial peptides
Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3
-
Stage composition beats extra supervision for biology reasoning models
How Post-Training Shapes Biological Reasoning Models
-
FM representations hold up on shifted cancer data
Probing, Fusion, and Trustworthiness: A Systematic Evaluation of Foundation Model Representations for Multimodal Cancer Analysis
-
Three labels balance cost and power in lipid metabolism models
Balancing label resolution and computational cost in dynamical models of lipid metabolism
-
Framework recovers both growth laws and noise from data
A likelihood-based framework for simultaneously learning both noise and growth dynamics using biologically-informed neural networks
-
Model ties HDV drop to fourfold HBV DNA production rise
Mathematical Modeling of HDV RNA, HBV DNA, and HBsAg Dynamics during Lonafarnib-Based Therapy: Insights from the LOWR HDV-1 Study
-
Small LLMs surpass GPT-4o on biomedical claims at low cost
Small LLMs for Biomedical Claim Verification: Cost-Effective Fine-Tuning, Structural Dataset Shortcuts, and Cross-Domain Generalization
-
Vanilla Transformer reaches SOTA on cell perturbation prediction
OCOO-T : A Simple and Scalable Virtual Cell Model for Transcriptional Perturbation Response Prediction
-
Feature library recovers parameters from time series data
Interpretable model-free inference of parametric variation across time-series data through large-scale feature extraction
-
Gradient descent on distograms designs epitope-specific nanobody CDRs
EasyNano: rapid epitope-targeted nanobody CDR design via differentiable distogram optimization with ESMFold2
-
PINNs show Michaelis-Menten PK model needs tissue data
Physics-Informed Neural Networks for Chemotherapy Pharmacokinetics: Benchmarking the Clinical Estimator and Exposing Parameter Identifiability
-
Physics losses raise OOD AUC of synthesizability GNN
Physics-Aware Auxiliary Losses Improve Out-of-Distribution Generalization of a GNN Synthesizability Filter
-
Metric choice reverses winner in drug-response models
The Metric Picks the Winner: Evaluation Choice Flips Model Rankings for Drug-Response Prediction in Unseen Chemistry
-
Viral proteins expose nativeness axis in protein model embeddings
Viral Proteins Reveal Geometry of Protein Language Models
-
EvAMs predict intervention effects via do-operator on parameters
A structural causal framework for interventions on evolutionary accumulation models
-
EvAMs gain do-operator interventions for target ranking
A structural causal framework for interventions on evolutionary accumulation models
-
ESMC autoencoder features predict enzyme functions at 79% accuracy
Interpretable enzyme function prediction via sparse autoencoder features of ESMC across the microbial protein universe
-
Span doubles early ctDNA warnings for indolent breast cancer progression
Seeing Below the Limit of Detection: A Censored-Poisson Bayesian Latent-Growth Change-Point Detector (the Span Detector) for Serial ctDNA in HR+/HER2- Metastatic Breast Cancer
-
Memory bank restores spectrum use in peptide sequencing
MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry
-
Magnitude turns cell point patterns into interaction features
Magnitude-Based Features for Multispecies Spatial Data
-
Hybrid VAE suggests RHP monomer mixes for protein stabilization
DeepRHP: A Hybrid Variational Autoencoder for Designing Random Heteropolymers as Protein Mimics
-
PolyILR aligns orthonormal coordinates with any tree for proportion data
Tree-Structured Orthonormal Decomposition of the Aitchison Simplex
-
Embeddings turn discrete biome maps into continuous predictors
Continuous biome representations from Earth observation embeddings
-
Probabilistic pretraining improves ADME predictions by 7-10%
Probabilistic Contrastive Pretraining for Multi-task ADME Property Prediction
-
Hill functions alone hit sharpness bound n/4
Sharpness characterizes Hill functions
-
Masked regression separates local and distributed signals in EEG
Spatially Masked Regression Reveals Local and Distributed Predictability in Electrophysiological Recordings
-
Single-timepoint NGS fails to predict osimertinib resistance above chance
OncoTraj: a public benchmark for longitudinal resistance prediction in EGFR-mutant non-small-cell lung cancer on osimertinib
-
Helical structure key to modeling DNA shear rupture
A kinetic model of shear-induced rupture of short dsDNA
-
PDE often chosen over true ABM for wound healing data
Spatial Model Selection and Uncertainty Quantification: Comparing Continuous and Discrete Wound Healing Models
-
Structured coupling boosts discrete flows for DNA and peptides
Flexible Flows for Biological Sequence Design
-
Inference-time augmentations raise PPG AF detection AUROC up to 8.5%
A Comprehensive Inference-Time Augmentation Framework in Physiological Signals: Application to PPG-Based AF Detection
-
Inference augmentations lift PPG AF detection AUROC by up to 8.5%
A Comprehensive Inference-Time Augmentation Framework in Physiological Signals: Application to PPG-Based AF Detection
-
YOLO on spectrograms doubles bird call localization accuracy
Time-frequency localization of bird calls in dense soundscapes
-
Modular health digital twins couple across scales via seven layers
OmniBioTwin: A System-of-Twinned-Systems Framework for Health Digital Twins