#research

Research

1100 stories tagged Research · page 10 of 46

researchJul 20, 2026

DrawingVQA: First AI Benchmark Tests Multimodal Models on Real-World Construction Drawings

Researchers introduced DrawingVQA, the first benchmark to evaluate multimodal large language models (MLLMs) on real-world construction drawings — a uniquely complex domain fusing abstract geometry, symbols, tables, and technical text. The benchmark uses 33 professional 'Issued for Construction' drawings and 92 expert-crafted questions to test AI's visual-textual reasoning in architecture and civil engineering.

via ArXiv cs.AI#AI #Benchmark #Engineering

researchJul 20, 2026

GraphDx: A Cost-Aware Multi-Agent AI Framework for Sequential Medical Diagnosis

Researchers introduced GraphDx, a multi-agent AI framework that uses a Medical Diagnosis Knowledge Graph to balance diagnostic accuracy with resource costs, reducing unnecessary testing in sequential medical diagnosis.

via ArXiv cs.AI#AI #Healthcare #Diagnosis

researchJul 20, 2026

Cura 1T: Specialized AI Model for Agentic Healthcare Tasks

Researchers introduced Cura 1T, a healthcare-specialized AI model trained via a human-gated self-evolution loop to handle patient consultations, clinical reasoning over text and images, interactive diagnosis, and EHR tool use without degrading performance across tasks.

via ArXiv cs.AI#Healthcare #Models #Ehr

researchJul 20, 2026

AnovaX: A Local, Multi-Agent Voice Assistant with LLM Planning and Adaptive Recovery

Researchers describe AnovaX, a local-first voice assistant that runs entirely on the user's computer, using an LLM planner (Gemini) to generate JSON plans of tool calls, a multi-agent orchestrator, and adaptive recovery — all without sending audio to the cloud.

via ArXiv cs.AI#Voice Assistant #Privacy #Research

$AI Reviewers Don't Always Improve Math Problem Solving — Peer Discussion Beats Structured Review on Hard Problems$

researchJul 20, 2026

AI Reviewers Don't Always Improve Math Problem Solving — Peer Discussion Beats Structured Review on Hard Problems

A new study on 4,181 math problems found that adding AI reviewers to multi-agent systems doesn't always improve accuracy. While reviewers help with harder problems, they don't boost results for easier ones. For the most difficult problems, a simple peer discussion method outperformed a structured planner-executor-reviewer pipeline.

via ArXiv cs.AI#AI #Math #Research

researchJul 20, 2026

LLMs as Unified Multimodal Learners for Clinical Prediction: Converting All Patient Data into Text

A new arXiv study proposes converting all patient data—clinical narratives, vital signs, lab values, and comorbidities—into a single natural language sequence and fine-tuning a pretrained LLM, eliminating the need for task-specific fusion architectures in clinical prediction.

via ArXiv cs.CL#AI #Healthcare #Research

generalJul 19, 2026

Study: Using AI Makes People Less Likely to Admit They Don't Know Something

A new study from the University of California reveals that relying on AI tools like ChatGPT or Claude can make people less willing to admit ignorance, leading to overconfidence and poor decision-making in areas like health and finance.

via Hacker News AI#AI #Research #Learning

researchJul 17, 2026

Introspection Fine-Tuning (IFT): Training Small LLMs to Detect and Report Internal Changes

Researchers introduce Introspection Fine-Tuning (IFT), a method that trains small language models to detect and report perturbations in their own internal activations. This breakthrough could make AI systems more reliable and transparent by enabling self-monitoring in smaller, efficient models.

via ArXiv cs.CL#AI #Research #Language Models

researchJul 17, 2026

Just Keep Prompting (JKP): New Framework Tests Vision-Language Model Stability Under Repeated Questioning

Researchers introduced Just Keep Prompting (JKP), a multi-turn evaluation framework that measures how well Vision-Language Models (VLMs) maintain epistemic stability when users repeatedly challenge, question, or contradict their answers across up to 10 follow-up turns.

via ArXiv cs.CL#AI #Research #Vision Language Models

researchJul 17, 2026

ArXiv Study: Information-Theoretic Limits Prove AI Reliability Has a Ceiling, Regardless of Scale

A new ArXiv paper proves that large language models (LLMs) have an inherent reliability ceiling that no amount of scaling can overcome. The study decomposes output uncertainty into a resolvable component (closable with more context) and a subjective component (inherent to task ambiguity), and shows that autoregressive generation further degrades this ceiling.

via ArXiv cs.CL#AI #Reliability #Research

researchJul 17, 2026

New Research Enhances Neuro-Symbolic AI with Probability for Smarter, More Transparent Reasoning

Researchers have enhanced neuro-symbolic AI by adding probability to logical reasoning, enabling AI to make educated guesses about unknown information while preserving its logical structure. This could lead to robots that learn and reason like humans, but with more transparency and structure.

via ArXiv cs.AI#AI #Neuro Symbolic #Research

researchJul 17, 2026

New Metric Measures AI Tool Efficiency Like a Report Card

Researchers created a way to measure how well AI tools work together. It helps AI assistants use the right tools at the right time, making them faster and more reliable.

via ArXiv cs.CL#AI #Tools #Research

researchJul 17, 2026

Theory-Level Autoformalization: AI That Formalizes Entire Theories, Not Just Isolated Statements

A new arXiv position paper from researchers argues for a shift in AI autoformalization from single statements to complete theories—including axioms, definitions, and lemmas—to create unified, machine-verifiable formal knowledge bases. This could transform how complex knowledge in mathematics, law, and science is structured and validated.

via ArXiv cs.AI#Research

researchJul 17, 2026

New AI Research Automatically Generates Task-Specific Prompt Guidelines to Improve LLM Accuracy

Researchers from ArXiv cs.CL have developed a system that automatically evolves prompt guidelines tailored to specific tasks, addressing the common problem of underspecified user queries. This could make large language models like ChatGPT and Claude more accurate and user-friendly without requiring prompt engineering expertise.

via ArXiv cs.CL#AI #Research #Prompt Engineering

researchJul 17, 2026

MAPS: New AI Framework Lets Agents Hold Conversations While Keeping Their Own Perspectives

Researchers introduced MAPS (Multi-Agent Perspective Spaces), a framework that enables multiple AI agents to maintain individualized beliefs, emotions, and cognitive styles during dialogue, avoiding the semantic uniformity of current systems. It uses domain-weighted profiles, GRU-based memory, and token-level attention for interpretable, diverse interactions.

via ArXiv cs.CL#AI #Research #Multi Agent

researchJul 17, 2026

AI Agents Lose Information When Communicating via Text, Study Finds

A new ArXiv study shows that LLM-based AI agents lose information when communicating via plain text, suggesting they possess an internal 'world model' that exceeds textual expressibility. This finding has implications for multi-agent system design and alignment.

via ArXiv cs.CL#AI Agents #Communication #Research

researchJul 16, 2026

SPINE: New AI Framework Automates Robot Deployment, Cutting Need for Expert Calibration

Researchers introduced SPINE (Scalable Physical Integration with ageNtic Expertise), an agentic AI framework that automates the debugging and deployment of bimanual robots, reducing the need for expert-driven calibration and making advanced robotics more accessible.

via ArXiv cs.AI#AI #Robotics #Embodied AI

researchJul 16, 2026

Networked Intelligence: Researchers Propose AI Teams to Solve Complex Scientific Problems

A new arXiv paper proposes 'networked intelligence' — teams of specialized AI systems collaborating like human scientists — to tackle complex scientific challenges that single models cannot solve alone.

via ArXiv cs.AI#AI #Research #Science

researchJul 16, 2026

New Mathematical Framework Enables Insurance for Autonomous AI Systems

A new study from ArXiv cs.AI introduces a mathematical framework for underwriting and pricing insurance tailored to autonomous AI systems. The model accounts for risks like decision-making, tool use, and environmental interactions, filling a gap left by traditional insurance policies.

via ArXiv cs.AI#Autonomous AI #Research #Safety

researchJul 16, 2026

Oracle Agent Memory: A Breakthrough for Long-Term AI Assistants

Researchers have developed a new system called Oracle Agent Memory to help AI assistants remember important details over long conversations. This could make AI tools more reliable for tasks that span multiple sessions, like complex projects or ongoing customer support.

via ArXiv cs.AI#AI #Memory #Research

researchJul 16, 2026

Self-Improving AI Agents: New Survey Explores How Autonomous Systems Learn and Adapt

A new survey on arXiv frames self-improving AI agents as adaptive systems that convert experience into capability gains with minimal human input. The research defines modern agents as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic.

via ArXiv cs.AI#AI #Research #Self Improving

researchJul 16, 2026

Interventional Grounding Audits: Black-Box Test Checks If LLM Reasoning Actually Uses Its Premises

Researchers introduce interventional grounding audits, a black-box method that substitutes predicates in LLM chain-of-thought reasoning to test whether conclusions genuinely depend on stated premises, evaluated on the ProntoQA benchmark.

via ArXiv cs.AI#AI #Research #Reasoning

researchJul 16, 2026

Safe-Psych Benchmark Tests Whether AI Models Ask Before Diagnosing in Psychiatry

Researchers introduced Safe-Psych, a sequential benchmark that evaluates how well large language models handle evolving diagnostic uncertainty in clinical psychiatry. Unlike existing medical benchmarks that assume complete information upfront, Safe-Psych tests whether models request clarification or abstain when data is insufficient.

via ArXiv cs.CL#AI #Healthcare #Research

researchJul 16, 2026

DROPJ: New AI Training Method Uses Human Justifications for Safer Agent Behavior

Researchers introduced DROPJ, a human-centered method that trains AI agents safely by combining world models with human feedback and justifications. This approach could make AI systems more reliable in safety-critical domains like healthcare and autonomous driving.

via ArXiv cs.AI#Safety #Reinforcement Learning #Human Feedback