AI/ML arXiv cs.AI

ART for Diffusion Sampling: Continuous-Time Control and Actor-Critic Learning

Introduces ART for Diffusion Sampling, a continuous-time control formulation that learns adaptive timesteps to improve sample quality in diffusion models.

AI/ML arXiv cs.AI

A Multi-Branch Hierarchy-Aware Framework for Heterogeneous Audio Classification

A new framework for heterogeneous audio classification using CLAP-based representations and hierarchy-aware classifiers to improve sound taxonomy accuracy.

AI/ML arXiv cs.AI

MolSight: A Graph-Aware Vision-Language Model for Unified Chemical Image Understanding

Introduction of MolSight, a graph-aware vision-language model that integrates molecular topology to improve the understanding of chemical images for drug discovery.

AI/ML arXiv cs.AI

Do Newer Lightweight CNNs Perform Better Under Resource Constraints? A Controlled Multigenerational Study of Architecture, Initialization, Training Budget, and Efficiency

A comparative study of lightweight CNNs revealing that newer architectures do not always outperform older ones like EfficientNet-B0 under resource constraints.

AI/ML arXiv cs.AI

Mirror Illusion Art

AutoMIA is an automated pipeline for creating 3D mirror illusion art by jointly optimizing shape and color from 2D images.

AI/ML arXiv cs.AI

Towards Load-Aware Prefill Deflection for Disaggregated LLM Serving

A load-aware prefill deflection scheduler for disaggregated LLM serving that reduces P95 TTFT by up to 81% by interleaving prefill and decode phases.

AI/ML arXiv cs.AI

OpenSafeIntent: Evaluating Intent-Calibrated Safe Completion Across Dual-Use Prompt Sets

OpenSafeIntent introduces a benchmark to evaluate LLM safety by varying intent while keeping tasks fixed, revealing brittle safety behaviors.

AI/ML arXiv cs.AI

SPLIT: Cross-Lingual Empathy and Cultural Grounding in English and Ukrainian LLM Responses

The SPLIT benchmark evaluates cross-lingual empathy and cultural grounding in LLMs, specifically comparing English and Ukrainian responses in crisis contexts.

AI/ML arXiv cs.AI

Beyond the Performance Illusion: Structure-Aware Stratified Partitioning and Curriculum Distributionally Robust Optimization for Spatially Correlated Domains

Proposed a framework (SASP and CDRO) to prevent data leakage and hidden stratification in AI evaluation for spatiotemporally correlated domains.

Software Engineering arXiv cs.AI

Prompt Coverage Adequacy

Introduction of 'Prompt Coverage Adequacy', a testing metric for LLM-generated code that measures how well test suites satisfy prompt requirements.

AI/ML arXiv cs.AI

SA-HGNN: Sample-Adaptive Hyperbolic Graph Neural Network for EEG-Based Depression Recognition

SA-HGNN is a hyperbolic graph neural network designed to recognize depression by capturing hierarchical functional connectivity in EEG brain networks.

Cybersecurity Hacker News

Soatok's Informal Guide to Threat Models

A guide by Soatok on threat modeling, discussing how to identify assets, threats, and mitigations for a system.

Software Engineering Hacker News

Show HN: A statically typed, cross-platform, easily bootstrappable build system

A presentation of a new statically typed, cross-platform build system designed to be easily bootstrappable.

AI/ML arXiv cs.AI

AIriskEval-edu: New Dataset for Risk Assessment in AI-mediated K-12 Educational Explanations

Introduction of AIriskEval-edu, a dataset for assessing pedagogical risks in AI-generated K-12 educational content.

AI/ML arXiv cs.AI

CausalSteward: An Agentic Divide-Conquer-Combine Copilot for Causal Discovery

CausalSteward (CAST) is a human-in-the-loop multi-agent framework for assembling large causal models using a divide-and-conquer approach.

AI/ML arXiv cs.AI

PhysMani: Physics-principled 3D World Model for Dynamic Object Manipulation

PhysMani is a framework that uses a physics-principled 3D Gaussian world model for better dynamic object manipulation in embodied AI.

AI/ML arXiv cs.AI

Conditional Co-Ablation: Recovering Self-Repair Backups in Transformer Circuits

Conditional Co-Ablation (CoAx) is a new method for discovering 'self-repair' backup circuits in Transformers, improving mechanistic interpretability.

AI/ML arXiv cs.AI

Robust for the Wrong Reasons: The Representational Geometry of LLM Robustness to Science Skepticism

Research on how LLMs respond to science skepticism, finding that they generally resist sycophancy but vary in their representational geometry.

AI/ML arXiv cs.AI

NeoMap: Training-free Novel-View Synthesis from Single Images and Videos

NeoMap is a training-free framework for high-fidelity novel-view synthesis from single images or videos using pre-trained video models.

Software Engineering arXiv cs.AI

Object Aligner: A Configurable JSON Schema Similarity Score for Graphs, Applied to LLM Prompt Optimization

Object Aligner is an open-source Python library for deterministically scoring JSON schema similarity, useful for LLM prompt optimization.