AI
MidTool: Mid-training Data Synthesis for Agentic Tool Use
Researchers have developed a new method for training large language models called MidTool. This approach focuses on the mid-training stage of model de...
AI
AI
Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimation
Researchers have proposed a method for efficiently allocating tasks to multiple AI models while considering the cost of estimating each model's perfor...
AI
AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement
Researchers have developed AI4AI-Bench, a benchmarking tool to evaluate the ability of large language models (LLMs) to design and improve their own tr...
AI
An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction
Researchers at McGill University propose a three-agent workflow for travel behavior modeling and weather-sensitive demand prediction. The approach int...
AI
A Virtual Member of a Community of Practice for the Society of Petroleum Engineers: From Prototype to Deployment
A virtual assistant called ATHENA has been developed to support the Society of Petroleum Engineers' Community of Practice. The system improves product...
AI
Transformer Models for Text Summarization: A Comparative Study of BART, BERT, and RoBERTa
A comparative study of three popular transformer-based models - BERT, RoBERTa, and BART - is presented in this article. The authors examine the archit...
AI
Automatic bioinformatic software named entity recognition from literature
Researchers have developed a new AI-powered tool called SNAIL that can automatically identify bioinformatics software and database names from scientif...
AI
Asymmetric Attention Heads: Structured Head-Wise Context Allocation for Transformer Attention
Researchers have proposed a new framework called Asymmetric Attention Heads (AAH), which allows different heads in the transformer attention mechanism...
AI
Hallucination as a Feature, not a Defect: Evaluating a multi-agent architecture to transform speculative language-model outputs into testable scientific hypotheses
Researchers have proposed a new multi-agent system to transform speculative language model outputs into testable scientific hypotheses. The system use...
AI
Towards On-Board Implementation of ML-Based Helicopter Weight Estimator
Researchers from Airbus have proposed a machine learning-based system to estimate the weight of helicopters during takeoff. The system uses data from ...
AI
Represented but Ignored: A Causal Account of Prosodic Underuse in Audio-Language Models
Researchers have found that audio-language models (LLMs) often preserve prosodic information from speech but struggle to use it effectively. A new met...
AI
Time-Series Retrieval for Grounding Multimodal Language Models in Remaining Useful Life
Researchers have proposed a framework to improve predictions of remaining useful life (RUL) for machines and equipment. They used large language model...