AI/ML arXiv cs.AI

GrocLM: Grocery Category Recommendation in E-Commerce with Large Language Models

GrocLM is a fine-tuned LLM using a two-stage LoRA strategy and trie-based decoding for grocery category recommendations in e-commerce.

AI/ML arXiv cs.AI

Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation

Crystalis is a framework that uses progressive nucleation and semantic annealing to help LLMs reliably produce structurally correct multi-view visualizations.

AI/ML arXiv cs.AI

PATHFinder Agent for Tailored Prenatal Care

PATHFinder Agent is a conversational agent designed to provide tailored prenatal care plans based on ACOG guidelines.

AI/ML arXiv cs.AI

LLM Scheming Inversely Scales with Pretraining Language Coverage

Research using the Petri auditing framework finds that LLM 'scheming' behaviors are more prevalent in low-resource languages than high-resource ones.

AI/ML arXiv cs.AI

ProcAgent: An Agentic Framework for Procedural Task Guidance on Edge with Human-in-the-Loop

ProcAgent is an on-device, vision-based procedural assistant that runs on NVIDIA Jetson AGX Orin to provide real-time adaptive guidance for physical tasks.

AI/ML Hacker News

Show HN: Lean4 Datalog DSL Based on Google Zanzibar for AI Projects

A new Domain Specific Language (DSL) for Lean4 based on Google Zanzibar, designed to facilitate the development of AI projects.

Other Ars Technica

Audi has a new flagship designed with the US in mind: The 2027 Q9

Audi announces the 2027 Q9, a new flagship full-size SUV specifically designed for the US market.

AI/ML arXiv cs.AI

SQBench: A Benchmark for Evaluating Task Delivery by Language-Model Agents in Production-Oriented Workflows

Introduction of SQBench, a benchmark for evaluating the ability of LLM agents to deliver verifiable deliverables in production-oriented workflows.

AI/ML arXiv cs.AI

AgentOmnia: Scaling Agentic Models for Full-Scenario Applications

AgentOmnia is a framework for scaling agentic models across diverse applications using a comprehensive taxonomy and a new benchmark called OmniaBench.

AI/ML arXiv cs.AI

CachedSearch: Training-Free Cached Exploration for Test-Time Search in Video Diffusion

CachedSearch introduces a training-free caching mechanism to accelerate test-time search in video diffusion models without significant quality loss.

AI/ML arXiv cs.AI

An Ontology for Machine Learning Interatomic Potentials

The MLIPs ontology provides a standardized OWL 2 DL framework to describe machine learning interatomic potentials, improving reproducibility in materials science.

AI/ML arXiv cs.AI

Characterisation of Density-based FM generation methods in the context of Information Fusion

Research on density-based Fuzzy Measure (FM) generation methods to improve information fusion and provide confidence intervals for fusion results.

AI/ML arXiv cs.AI

CAPT: A Multi-task Continuous Autoregressive Transformer enabling Cross-dataset and Cross-species Transfer for Calcium Population Dynamics

CAPT is a continuous autoregressive transformer designed for calcium population dynamics that generalizes across different datasets and species.

AI/ML arXiv cs.AI

SeekJudge: A Practical Reward Framework for Reinforcement Learning in Computer-Use Agents

SeekJudge is a practical reward framework using specialized agents to provide accurate and efficient rewards for RL in computer-use agents.

AI/ML arXiv cs.AI

TopoFE: topology-aware LLM-guided Automated Feature Engineering

TopoFE is a topology-aware evolutionary framework that uses LLMs to automate feature engineering for tabular data more efficiently.

Other Hacker News

Does every question mark deserve a Betteridge?

A discussion on the 'Betteridge's law of headlines', suggesting that any question mark in a headline implies a negative answer.

Software Engineering Hacker News

Hooray for the Sockets Interface

A technical appreciation of the Sockets interface and its role in network programming.

Other Hacker News

Beyond Greece and Rome

An exploration of intellectual and cultural history beyond the classical Greek and Roman traditions.

Tech Business/VC Hacker News

Chip stocks slide in US and Asia as AI jitters rattle investors

Chip stocks are declining in the US and Asia due to investor concerns regarding AI growth and sustainability.

AI/ML arXiv cs.AI

ConsistencyGate: Preventing Memory Contamination in LLM Agents via Self-Consistency Admission Control

Introduces ConsistencyGate, a write-time admission control mechanism to prevent hallucinated facts from contaminating the external memory of LLM agents.