AI Chatbots for Teen Mental Health Show Promise but Evidence Remains Thin, Major Review Finds
A new PRISMA-ScR scoping review of 19 studies finds that AI chatbots and machine learning tools for adolescent mental health ...
Tag: Large Language Models
ChatGPT and Gemini Face a Histology Test: How Well Can AI Chatbots Spot Basal Cell Carcinoma Under the Microscope?
A UCLA research team tested whether ChatGPT and Gemini could correctly identify basal cell carcinoma from histologic images, revealing both ...
AI Sentiment Analysis Alone Cannot Tell Truth From Health Misinformation, Major Review Finds
A systematic review of thirty studies finds that sentiment analysis can flag emotional tone in health content but cannot determine ...
AI Reads Brain Waves and Explains Itself: New Framework Turns EEG Signals Into Plain-Language Depression Clues
A new evidence-guided framework called E-MDD combines deep EEG representations, neurophysiological biomarkers, and a constrained large language model to detect ...
How We Measure Recommenders Matters: Rethinking Evaluation for Real-World Search and Recommendation
A new Frontiers in Big Data editorial argues that recommender and search systems must be evaluated with robust, responsible and ...
Your Old Windows 10 PC Could Become a Free Private AI Server
Researchers transformed a retired Windows 10 workstation with a consumer GPU into a private AI server that beats mid-range and ...
Why AI Still Struggles to Answer Medical Questions in Bangla
A new review finds that Bangla medical question answering lags far behind English systems due to dataset scarcity, translated corpora, ...
AI Chatbots Flunk the Heart Failure Test: Even the Best Hallucinated in Landmark Patient Education Study
A new cross-sectional evaluation of ChatGPT, DeepSeek, Perplexity, and Microsoft Copilot found that all four AI chatbots produced hallucinated heart ...
NASA-built AI assistant taps verified aviation tools to cut hallucination risk
NASA Ames researchers built an LLM-based aviation decision-support system that routes natural language queries through nine validated expert models, achieving ...
Retrieval Beats Fine-Tuning: Teaching AI the Science of 3D Printing
A new study finds retrieval-augmented generation dramatically outperforms fine-tuning when adapting large language models to expert-level additive manufacturing questions.
Investment Advisers Reveal AI as Risk More Often Than Capability
A validated large-scale audit of SEC advisory filings finds AI disclosed as a risk more often than as a capability, ...
AI Agents Now Draft Floor Plans That Pass Municipal Building Codes
A multiagent AI framework called LayoutGen couples large language model creativity with deterministic geometric verification, cutting municipal building code violations ...
AI System Teaches Robots to Dismantle Dead Batteries With Human Help
A new AI planning system combines a single-call language model, a verifier, and a deterministic grounding filter to help robots ...
Why Science Communication Is Breaking Down, and How Philanthropy Can Fix It
A Milken Institute team argues that the biomedical communication pipeline is breaking down at publication, distribution, and public engagement, and ...
ARTEMIS turns thousands of papers into reusable AI memory for regulatory science
FDA researchers developed ARTEMIS, a framework that distills thousands of monoclonal antibody papers into compact, steerable AI memory that outperforms ...
AI Reviewers Match Human Experts in Landmark Scoping Review Test
Artificial intelligence has been creeping into the machinery of science for years, but a new benchmark from Johns Hopkins researchers ...
AI Checklist Helps Radiologists Read Tricky Heel Surgery X-Rays Faster and More Accurately
A multicenter study of six radiologists found that a pre-generated GPT-5 checklist and draft report raised binary escalation accuracy on ...
AI Learns to Read Between the Lines: LLM-Powered Recommender Boosts Accuracy While Keeping Data Private
A new hybrid framework called LLMTransRec combines transformer sequence modeling with large language model embeddings to improve recommendations across sparse ...
New AI Shield MIRAGE Fights Hidden Attacks on Retrieval-Augmented Language Models
Researchers have unveiled MIRAGE, a two-stage adaptive defense framework that uses embedding-space density analysis and attack-adaptive knowledge integration to protect ...
Cancer Patients Are Turning to Chatbots: Nurses Warn the Evidence Is Not Ready
A new correspondence in Supportive Care in Cancer urges caution as patients and caregivers increasingly turn to generative AI tools ...
POPULAR NEWS
-
One in Five Adolescents in Asian Birth Cohort Engaged in Repeated Self-Harm by Age 13
-
Baicalein Identified as Direct IDO1 Inhibitor in Ulcerative Colitis
-
Indicator-Based Review Assesses Policy Gaps in Iran’s Brucellosis Control Program
-
Humic Substances Boost Passion Fruit Seedling Vigor Only Alongside Mineral Fertilization