AI Agents That Debate Grades: New System Scores Essays Trait by Trait Without Training
A new multi-agent AI system called MATE uses specialized debating agents and a coordinating agent to score essays across multiple ...
Tag: large language models
AI screens 52,923 EU research projects to map transport innovation in days
European Commission researchers used large language models to automatically screen and classify EU-funded transport research projects, cutting curation work from ...
LLM Framework Maps an Uneven Evidence Base on Air Pollution and Cognition
A human-validated LLM-assisted role-analysis framework found that the post-2020 air pollution and cognition literature is volume-rich but structurally uneven.
Research report: Low-resource hallucination detection in LLMs on multi-task datasets via iterative pseudo-labeling using confidence thresholding and active learning
Researchers report a semi-supervised framework combining pseudo-labeling, active learning, and ensemble methods that detects fluent hallucinations in large language model ...
Effect of guideline-assisted prompting on large language model performance in medication-related osteonecrosis of the jaw–related endodontic decision-making
Supplying the 2022 AAOMS position paper in prompts significantly improved Claude's performance on MRONJ-related endodontic scenarios, erasing its baseline gap ...
AI Reads Student Records to Predict English Proficiency in Vocational Colleges
A new interpretable AI framework combining large language model embeddings with structured educational data predicted vocational college students' English proficiency ...
Research report: Paired response final divergence signatures of jailbreak vulnerability in large language models
A new representational analysis of 15 open-source language models shows that jailbreak vulnerability leaves measurable, family-specific traces in hidden states, ...
Machine learning models flag higher cancer risk from routine health data
Machine learning models using routine health data showed potential to identify people at higher risk of several common cancers, with ...
Self-Taught AI Model Cuts Speech Translation Errors While Tripling Speed
A new model called M-SREC pairs a lightweight Mistral backbone with self-supervised noise simulation to cut spoken translation error-correction latency ...
AI Reads Police Logs to Count Vulnerability, But Only With Heavy Human Correction
A study of nearly 3,000 UK police incident logs shows fine-tuned language models can estimate vulnerability prevalence, but only after ...
-
Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study
-
University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept
-
Bee body mass, pathogens and local climate influence heat tolerance
-
Researchers record first-ever images and data of a shark experiencing a boat strike
-
Groundbreaking Clinical Trial Reveals Lubiprostone Enhances Kidney Function
Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!