Work Force

Training-Free Generative Sampling via Moment-Matched Score Smoothing

Diffusion models generate samples by denoising along the score of a perturbed target distribution. In practice, one trains a neural diffusion model, which is computationally expensive. Recent work suggests that score matching implicitly smooths the empirical score, and that this smoothing bias promotes generalization by capturing low-dimensional data geometry. We propose moment-matched score-smoot...

STRIDE: Training-Free Diversity Guidance via PCA-Directed Feature Perturbation in Single-Step Diffusion Models

Distilled one-step (T=1) or few-step (T$\leq$4) diffusion models enable real-time image generation but often exhibit reduced sample diversity compared to their multi-step counterparts. In multi-step diffusion, diversity can be introduced through schedules, trajectories, or iterative optimization; however, these mechanisms are unavailable in the few-step or single-step setting, limiting the effecti...

When Policy Entropy Constraint Fails: Preserving Diversity in Flow-based RLHF via Perceptual Entropy

RLHF is widely used to align flow-matching text-to-image models with human preferences, but often leads to severe diversity collapse after fine-tuning...

Mechanisms Matter: Transportability of Cellular Perturbation Effects

Predicting cellular responses to genetic or chemical perturbations across biological contexts is central to drug development and disease understanding...

Portable Active Learning for Object Detection

Annotating bounding boxes is costly and limits the scalability of object detection. This challenge is compounded by the need to preserve high accuracy...

Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance

We present EDDY (Exact-marginal Diversification via Divergence-free dYnamics), a guidance mechanism for diffusion and flow matching models that promot...

Machine learning approaches for the identification and analysis of enterotoxin genes in Staphylococcus aureus genomes

Staphylococcus aureus produces a broad range of enterotoxins that act as superantigens, disrupting host immune responses and resulting in a myriad of ...

Functional Profiling of Thousands of Sequence-Diverse Protease Homologs with GROQ-seq

High-quality datasets that span broad sequence diversity are essential for understanding protein sequence-function relationships beyond local mutation...

Towards Open World Sound Event Detection

Sound Event Detection (SED) plays a vital role in audio understanding, with applications in surveillance, smart cities, healthcare, and multimedia ind...

Fidelity, Diversity, and Privacy: A Multi-Dimensional LLM Evaluation for Clinical Data Augmentation

The scarcity of high-quality annotated medical data, particularly in mental health, poses a significant bottleneck for training robust machine learnin...

Disrupted oral microbial networks and reproducible community signatures implicate the oral-gut axis in Crohn's disease

Background: Emerging evidence suggests that the oral microbiome may contribute to aberrant gut immune responses in Inflammatory Bowel Disease (IBD). M...

ACPO: Anchor-Constrained Perceptual Optimization for Diffusion Models with No-Reference Quality Guidance

Diffusion models have achieved remarkable success in image generation, yet their training is predominantly driven by full-reference objectives that en...

Development and validation of a lesion-supervised deep learning system for diabetic retinopathy grading according to UK national screening criteria

Background: Diabetic retinopathy (DR) is the leading cause of preventable blindness among working-age adults worldwide, yet screening coverage remains...

Diverse Image Priors for Black-box Data-free Knowledge Distillation

Knowledge distillation (KD) represents a vital mechanism to transfer expertise from complex teacher networks to efficient student models. However, in ...

Improving Diversity in Black-box Few-shot Knowledge Distillation

Knowledge distillation (KD) is a well-known technique to effectively compress a large network (teacher) to a smaller network (student) with little sac...

Meta-Ensemble Learning with Diverse Data Splits for Improved Respiratory Sound Classification

Training reliable respiratory sound classification models remains challenging due to the limited size and subject diversity of datasets. Ensemble meth...

Zero-to-CAD: Agentic Synthesis of Interpretable CAD Programs at Million-Scale Without Real Data

Computer-Aided Design (CAD) models are defined by their construction history: a parametric recipe that encodes design intent. However, existing large-...

PlankFormer: Robust Plankton Instance Segmentation via MAE-Pretrained Vision Transformers and Pseudo Community Image Generation

Plankton monitoring is essential for assessing aquatic ecosystems but is limited by the labor-intensive nature of manual microscopic analysis. Automat...

Soft Label Pruning and Quantization for Large-Scale Dataset Distillation

Large-scale dataset distillation requires storing auxiliary soft labels that can be 30-40x larger on ImageNet-1K and 200x larger on ImageNet-21K than ...

Geometry-Guided 3D Visual Token Pruning for Video-Language Models

Multimodal large language models have demonstrated remarkable capabilities in 2D vision, motivating their extension to 3D scene understanding. Recent ...