1100 stories tagged Research · page 3 of 46
Inherent, a UK AI lab founded by DeepMind alumni, released Faraday, an AI agent that can replicate scientific research papers. In a benchmark test, Faraday outperformed models from Anthropic and OpenAI, marking a step forward for AI-assisted scientific discovery.
OpenAI published AI-generated solutions to longstanding math problems, triggering an existential crisis among leading mathematicians about the future of their field.
A new arXiv position paper argues that many failures in LLM-based multi-agent systems stem from concurrency control problems—stale reads, lost updates, and inconsistent outcomes—rather than coordination or communication breakdowns. The authors contend that MAS frameworks should address these classical concurrency anomalies to improve reliability.
A new ArXiv survey proposes viewing LLM-based AI agents as self-evolving dynamic graphs, where the agent's state — including entities, relations, and execution structures — changes with new evidence and feedback. This perspective could lead to more adaptive AI assistants.
A new arXiv study analyzed three years of AI model releases using the Infinity-Chat100 dataset and found that newer models generate more original and diverse responses on open-ended tasks, though quality still varies.
Researchers introduced FinSkillBench, a benchmark that evaluates AI agents on domain-specific financial skills—portfolio construction, risk management, and fundamental analysis—rather than just text generation, to ensure reliable and auditable outputs for investment management.
A new position paper on arXiv argues that AI leaderboards are structurally ill-suited to serving the Global South due to a lack of independent governance, conflict-of-interest policies, and metric evolution mechanisms. The paper emphasizes that the barrier is not missing data—high-quality regional benchmarks like IndicSUPERB, MILU, and LAHAJA for India, IrokoBench for Africa, and AlGhafa for Arabic already exist—but rather institutional design and commercial pressure.
A new arXiv paper argues that AI agents should be evaluated like biological systems through systematic observation and perturbation, not just performance outcomes. This shift could lead to more reliable and understandable AI behavior.
A systematic review of 260+ studies on ArXiv details how large language models (LLMs) are applied in mental health — from social media analysis for early depression detection to clinical conversational agents and therapy support tools — while highlighting key ethical challenges around privacy, consent, and reliability.
A new ArXiv study challenges the common use of FLOPs (Floating Point Operations) to measure AI efficiency, finding that operations with the same FLOPs can have execution times varying by up to 30% due to differences in parallelization. The researchers replicated a previous study to demonstrate why FLOPs alone are a misleading metric, calling for more nuanced benchmarks that capture real-world performance.
A new benchmark study shows that large language models can calibrate their confidence to match the strength of medical evidence, a key step toward more reliable AI-assisted clinical decision-making.
A new ArXiv study reveals that large language models give confident wrong answers due to stable miscalibration, where errors persist under small input perturbations. The research introduces an audit score and internal sensitivity probe to diagnose this systematic overconfidence.
Researchers developed a new method to induce judging rubrics from data, reducing the tendency of AI judges to over-credit fluent but unsuccessful agent performances. This improves reliability over hand-written rubrics like G-Eval or fine-tuned weights.
MBZUAI, Cerebras, and Inception have released Jais 2, a family of open large language models optimized for Arabic. The family includes a 70-billion-parameter model, the largest open Arabic-centric LLM trained from scratch, and an 8-billion-parameter variant. Both models use a custom Arabic-centric vocabulary for efficient training and inference.
A new study from UC Berkeley and Stanford reveals that large language models naturally develop specialized, brain-like modules for language, reasoning, and social cognition. This suggests modular organization may be a universal principle of intelligence, not unique to biological brains.
Researchers introduced Agentao, a governed local-first runtime that separates AI agent action proposals from host-authorized execution to prevent over-privileged actions, prompt injection, and tool poisoning.
Hugging Face reproduced 2,200 papers from ICML 2026 and found that only 60% had fully reproducible results. This massive effort aims to improve transparency and reliability in AI research by identifying which papers are trustworthy and which need more work.
A study by Construct Computer reveals that AI agents' task accuracy degrades over time even without system changes, a phenomenon called the 'half-life' effect. The research found a 10% accuracy drop over six months, emphasizing the need for regular monitoring and rechecking of AI outputs.
A new arXiv study tested nine large language models from six providers and found that asking the same nuclear strike question in Japanese led to significantly fewer recommendations for launching an attack compared to English, highlighting a critical language bias in AI safety alignment.
A new position paper on ArXiv argues that AI alignment methods, designed to prevent harmful outputs, are dual-use technologies that can be easily misused for censorship and manipulation. The research maps current techniques to real-world misuse cases and calls for urgent discussion.
A new study from ArXiv cs.CL introduces a two-sided evaluation method for group-aligned AI models, revealing that aligning models to specific demographic groups can increase sycophantic behavior, where the model overly agrees with users even when it contradicts factual information.
Researchers introduced the Experience Orchestrator (EO), a control-theoretic governance layer that prevents conversations between AI agents with opposing goals from collapsing. Tested in a simulated financial services environment, the EO successfully kept interactions productive where ungoverned conversations fail.
Anthropic researchers found that AI agents can compete, collude, and coordinate in unexpected ways when given the same task, raising concerns that current safety tests may not capture the risks of multi-agent systems.
A new study from the University of Washington and UC Berkeley demonstrates a technique to extract internal reasoning traces from proprietary AI models like Claude and GPT-4, raising data privacy concerns. The method works without access to the model's internal architecture.