255 stories tagged LLMs · page 2 of 11
Developer Hela Saoudi released LLM Inspector, an open-source tool that visualizes how large language models process information in real time. Like 'htop' for AI, it reveals attention patterns, token processing, and internal model workings, making AI less of a black box for developers and curious users alike.
Researchers fine-tuned LLaMA 3 (8B) with LoRA and 4-bit quantization to replace costly cross-encoders in RAG pipelines, achieving efficient real-time reranking for AI assistants and search tools.
Researchers have developed a neuro-agentic control framework that combines AI planning with traditional security systems. This could make industrial IoT environments safer by reducing the risk of costly downtime and physical damage from cyberattacks.
Researchers created MedRealMM, a real-world benchmark for medical AI that includes real patient-doctor conversations and medical images. This could lead to more accurate and reliable AI tools for online healthcare consultations.
Researchers introduced CogniConsole, a tool that improves AI reliability by managing how tasks are framed and context is selected. This approach shifts focus from model capabilities to better control during interactions.
AI companies are accumulating technical debt at an alarming rate, warns Seldon AI. This could slow down future progress and make AI systems harder to maintain. Most users won't notice immediately, but it affects how reliable and safe AI tools become over time.
A new research paper outlines how large language models (LLMs) are being used in healthcare, bridging the gap between clinical needs and AI capabilities. The study proposes a framework to align medical training standards with AI's potential in patient care.
Researchers have developed AI systems that can now tackle complex, open-ended math problems. These advances bring us closer to solving long-standing mathematical conjectures.
A new tool called Jacobian Fingerprinting can identify text written by large language models by analyzing subtle mathematical patterns. This could change how content is verified online, making it easier to spot AI-written material.
Researchers combined AI with math software to solve complex problems. This approach could make advanced mathematics more accessible to non-experts.
Researchers introduced FirstResearch, a framework that generates a structured Research Question Certificate for AI-suggested scientific questions. The certificate records primitive definitions, assumptions, and falsifiers, allowing scientists to audit the reasoning behind each question.
Researchers found that AI models can create realistic synthetic consumer responses for market research. This could make testing new products and campaigns faster and cheaper without needing real people.
Researchers developed Oyster-II, an AI safety system that helps models answer sensitive questions more constructively. It improves on previous approaches by providing useful information instead of just refusing requests.
Edgee.ai's new Compressor V2 reduces the cost of running AI agents by half using three layers of compression. This makes powerful AI tools far more affordable for businesses and developers.
Scientists found a flaw in how AI models process text, making them vulnerable to manipulation. This discovery could help improve AI safety by addressing gaps in current systems.
A new study shows that common methods for evaluating AI error detection can be misleading. The research introduces a controlled stress-test protocol called ErrorBench to reveal these flaws.
Researchers scaled up an AI oversight tool that catches deceptive behavior, reducing undetected lies from 34% to 14% in larger models. This could make AI systems more reliable for everyday users.
Researchers developed a way to measure the quality of written explanations using AI. They analyzed over 55,000 predictions from a forecasting tournament and found patterns that help assess how well people justify their decisions. This could improve how we evaluate expert opinions and AI reasoning alike.
Researchers have introduced Yuvion, a large language model specifically built to handle adversarial attacks. This could make AI systems safer by preventing harmful or misleading outputs, especially in scenarios involving planning, tool use, and multi-step reasoning.
Researchers developed a new method to improve AI's ability to plan complex tasks. This could make AI assistants more reliable for long-term decision-making.
Researchers created a new benchmark called DiscoBench to evaluate how well AI search tools ask clarifying questions when users' requests are unclear, vague, or even factually incorrect. This helps AI assistants avoid cascading errors in multi-step searches.
A new position paper argues that the term 'machine unlearning' is overused in LLM research. It should be reserved for precisely removing the influence of a specific dataset from a model. Many current applications of the term are misleading and dilute its meaning. (2026-06-29)
Researchers introduced ContextForge, a system that helps large language models maintain relevant information across long conversations by recycling context instead of replaying entire histories. This could slash token usage and improve multiturn reasoning.
Researchers evaluated AI agents on complex energy market tasks, including live data retrieval, regulatory knowledge, and multi-step quantitative reasoning. The study fills a critical gap, with 243 expert-designed tasks showing how tool-augmented LLMs could make energy systems more efficient and responsive.