Machine Learning

[R] VOID: Video Object and Interaction Deletion (physically-consistent video inpainting)

We present VOID, a model for video object removal that aims to handle *physical interactions*, not just appearance. Most existing video i...

FLUX 2 Pro (2026) Sketch to Image

I sketched a cow and tested how different models interpret it into a realistic image for downstream 3D generation, turns out some models ...

Improving AI models’ ability to explain their predictions

[2603.25024] Improving Infinitely Deep Bayesian Neural Networks with Nesterov's Accelerated Gradient Method

Abstract page for arXiv paper 2603.25024: Improving Infinitely Deep Bayesian Neural Networks with Nesterov's Accelerated Gradient Method

[2603.25112] Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory

Abstract page for arXiv paper 2603.25112: Do LLMs Know What They Know? Measuring Metacognitive Efficiency with Signal Detection Theory

[2603.24946] MobileDev-Bench: A Comprehensive Benchmark for Evaluating Language Models on Mobile Application Development

Abstract page for arXiv paper 2603.24946: MobileDev-Bench: A Comprehensive Benchmark for Evaluating Language Models on Mobile Application...

[2603.25109] MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness

Abstract page for arXiv paper 2603.25109: MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness

[2603.24917] Estimating near-verbatim extraction risk in language models with decoding-constrained beam search

Abstract page for arXiv paper 2603.24917: Estimating near-verbatim extraction risk in language models with decoding-constrained beam search

[2603.25099] Large Language Models as Optimization Controllers: Adaptive Continuation for SIMP Topology Optimization

Abstract page for arXiv paper 2603.25099: Large Language Models as Optimization Controllers: Adaptive Continuation for SIMP Topology Opti...

[2603.25083] Learning domain-invariant features through channel-level sparsification for Out-Of Distribution Generalization

Abstract page for arXiv paper 2603.25083: Learning domain-invariant features through channel-level sparsification for Out-Of Distribution...

[2603.25063] TopoPilot: Reliable Conversational Workflow Automation for Topological Data Analysis and Visualization

Abstract page for arXiv paper 2603.25063: TopoPilot: Reliable Conversational Workflow Automation for Topological Data Analysis and Visual...

[2603.25056] The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Creates Exploitable Vulnerabilities

Abstract page for arXiv paper 2603.25056: The System Prompt Is the Attack Surface: How LLM Agent Configuration Shapes Security and Create...

[2603.24764] Synthetic Cardiac MRI Image Generation using Deep Generative Models

Abstract page for arXiv paper 2603.24764: Synthetic Cardiac MRI Image Generation using Deep Generative Models

[2603.25052] Closing the Confidence-Faithfulness Gap in Large Language Models

Abstract page for arXiv paper 2603.25052: Closing the Confidence-Faithfulness Gap in Large Language Models

[2603.24752] Autotuning T-PaiNN: Enabling Data-Efficient GNN Interatomic Potential Development via Classical-to-Quantum Transfer Learning

Abstract page for arXiv paper 2603.24752: Autotuning T-PaiNN: Enabling Data-Efficient GNN Interatomic Potential Development via Classical...

[2603.25006] Improving Fine-Grained Rice Leaf Disease Detection via Angular-Compactness Dual Loss Learning

Abstract page for arXiv paper 2603.25006: Improving Fine-Grained Rice Leaf Disease Detection via Angular-Compactness Dual Loss Learning

[2603.24989] Learning Rollout from Sampling:An R1-Style Tokenized Traffic Simulation Model

Abstract page for arXiv paper 2603.24989: Learning Rollout from Sampling:An R1-Style Tokenized Traffic Simulation Model

[2603.24986] Rethinking Health Agents: From Siloed AI to Collaborative Decision Mediators

Abstract page for arXiv paper 2603.24986: Rethinking Health Agents: From Siloed AI to Collaborative Decision Mediators

[2603.24705] Amortized Inference for Correlated Discrete Choice Models via Equivariant Neural Networks

Abstract page for arXiv paper 2603.24705: Amortized Inference for Correlated Discrete Choice Models via Equivariant Neural Networks

[2603.24704] Conformal Selective Prediction with General Risk Control

Abstract page for arXiv paper 2603.24704: Conformal Selective Prediction with General Risk Control

[2603.24654] Spectral methods: crucial for machine learning, natural for quantum computers?

Abstract page for arXiv paper 2603.24654: Spectral methods: crucial for machine learning, natural for quantum computers?

[2603.24652] Demystifying When Pruning Works via Representation Hierarchies

Abstract page for arXiv paper 2603.24652: Demystifying When Pruning Works via Representation Hierarchies

[2603.24940] Evaluating adaptive and generative AI-based feedback and recommendations in a knowledge-graph-integrated programming learning system

Abstract page for arXiv paper 2603.24940: Evaluating adaptive and generative AI-based feedback and recommendations in a knowledge-graph-i...