Tag: large language models

This analysis explores the efficacy of large language models (LLMs) in assessing urinary system histology for medical education, drawing insights from research published in Scientific Reports. The evaluation focuses on the performance, potential, and limitations of LLMs in this specialized educational context.

Tilde AI has launched TildeOpen LLM, a groundbreaking open-source large language model with over 30 billion parameters. This model is specifically designed to support a wide array of European languages, addressing a critical gap in current AI offerings and promoting digital sovereignty within the EU.

This article explores the integration of Large Language Model (LLM) agents for automating Atomic Force Microscopy (AFM) experiments. It introduces the Artificially Intelligent Lab Assistant (AILA) framework and the AFMBench evaluation suite, detailing the challenges and successes in applying LLM agents to complex scientific workflows, from experimental design to data analysis.

This analysis delves into the groundbreaking development of the Dual-Inf framework, which significantly enhances the accuracy and explainability of differential diagnoses generated by large language models (LLMs). By creating a specialized dataset and employing a novel dual-inference approach, researchers are paving the way for more reliable AI-assisted clinical decision-making.

New research reveals that as few as 250 poisoned documents can create backdoor vulnerabilities in large language models, regardless of their size or training data volume. This finding challenges the long-held assumption that attackers need significant control over training data, suggesting data-poisoning attacks may be more feasible and accessible than previously believed.

This report synthesizes ten critical insights from the AMTA 2025 conference, exploring the evolving landscape of AI in translation. Key themes include the interdependence of Machine Translation (MT) and Large Language Models (LLMs), the necessity of domain adaptation for quality, the rise of multi-agent systems, and the persistent challenges of bias, safety, and evaluation. The article underscores the continued centrality of human linguists in the AI-driven translation ecosystem.

Researchers are leveraging large language models (LLMs) to dissect the complex web of human motivations in strategic situations. By analyzing how LLMs respond to carefully crafted prompts, scientists are gaining unprecedented insights into the underlying drivers of human behavior in economic games and beyond.

Large Language Models (LLMs) are prone to generating false information, a phenomenon known as "hallucination." This analysis explores the underlying causes, from training methodologies that reward guessing over accuracy to the inherent limitations of current AI architectures. It delves into the challenges of mitigating these "lies" and questions whether a fundamental shift in LLM training and evaluation is necessary to foster true reliability.

New research from Anthropic reveals that a mere 250 malicious documents can compromise large language models, regardless of their size, challenging long-held assumptions about AI security and data integrity.

Explore the fundamentals of Large Language Models (LLMs) in this instructional guide. Understand what LLMs are, how they function through prediction and transformer architectures, and their diverse applications across industries. Learn about their benefits, limitations, and the future of this transformative AI technology.

A groundbreaking study reveals pervasive age and gender distortions in online content and large language models, demonstrating how algorithms amplify these biases and impact real-world perceptions and opportunities.

A recent randomized study published in Nature Medicine highlights the significant potential of large language models (LLMs) to assist clinicians in patient care. While LLMs demonstrate impressive capabilities, future research must focus on understanding the cognitive processes underlying clinical reasoning to optimize their utility and ensure reliability.

Large language models (LLMs) are increasingly integrated into scientific workflows, prompting critical discussions about their implications. This analysis explores diverse perspectives on how LLMs should shape scientific practice, from research collaboration and data integrity to ethical considerations and the future of scientific discovery.

Explore the inner workings of Anthropic's Claude, a large language model, through a technical tutorial. This article delves into its multilingual capabilities, planning mechanisms in poetry, sophisticated mental math strategies, multi-step reasoning, and the intricacies of hallucination and jailbreaks, offering an instructional guide for understanding AI cognition.

Chinese researchers have achieved a world-first by using the Origin Wukong quantum computer to fine-tune a billion-parameter AI model. This breakthrough demonstrates significant improvements in training effectiveness and accuracy, paving the way for more efficient AI development and addressing "computing power anxiety." The hybrid approach leverages quantum principles to enhance classical AI models.