All 2,608 AI stories, newest first · page 100 of 109
A new paper introduces a mathematical framework for how environments can act as memory for agents in reinforcement learning. Experiments show that spatial paths can reduce the information needed to represent history.
A new study reveals a vulnerability in diffusion-based language models (dLLMs) that allows attackers to override safety mechanisms. The exploit achieves a 76.1% attack success rate on HarmBench, challenging assumptions about model alignment.
A new study analyzes how bias mitigation impacts the embedding spaces of BERT and Llama2, showing reduced gender-occupation disparities. The findings highlight measurable improvements in model fairness through representational analysis.
Researchers introduce RAMP, a novel approach combining reinforcement learning and action model learning for numeric domains. This method enables online learning directly from environment interactions, eliminating the need for pre-recorded expert traces.
Researchers introduce PilotBench, a benchmark for testing LLMs on flight trajectory and attitude prediction with safety constraints. The dataset includes 708 real-world general aviation trajectories with synchronized telemetry data.
Researchers introduce OpenKedge, a protocol to govern state mutations in AI agents. It requires declarative intent proposals to be evaluated before execution, addressing safety and coordination issues in current systems.
OpenClaw's latest update focuses on stability, audio transcription, and memory improvements. The release also includes fixes for Telegram approval deadlocks and timezone issues in dreaming.
OpenAI has released a comprehensive guide on responsible AI use, emphasizing safety, accuracy, and transparency. The guide offers best practices for users of tools like ChatGPT.
A new open-source tool audits websites for AI search engines, potentially disrupting the SEO industry. The tool offers capabilities previously sold for thousands of dollars per month.
A new survey and benchmark reveal that while large language models excel at medical exams, their clinical reasoning falls short. The study underscores the need for robust medical reasoning capabilities in real-world healthcare settings.
A new arXiv paper explores how agentic language models can enhance the creation of planning domains from natural language descriptions. The study highlights the limitations of current LLMs in this task and proposes a feedback framework to improve results.
Researchers developed a mathematical model to analyze how generated text shapes the public record, identifying two key forces: drift and selection. This work provides insights into the long-term evolution of language in AI systems.
A new study reveals that AI-driven personalization in marketing can maintain performance over time without constant human oversight. The research highlights the effectiveness of agentic systems in consumer applications.
Researchers introduce LOM-action, a framework that simulates business events within an ontology-driven graph to ground AI decisions in real-world scenarios. This approach promises more auditable and contextually relevant enterprise AI systems.
Researchers propose a domain-agnostic method using Hypergraph Neural Networks to accelerate Minimal Unsatisfiable Subset (MUS) enumeration, addressing the exponential search space challenge in constraint satisfaction problems. This approach avoids reliance on explicit variable-constraint relationships, broadening its applicability.
Google has announced significant updates to its AI models and tools, including new features for Gemini and enhanced capabilities for developers. These advancements aim to improve productivity and creativity across various applications.
Cloudflare has integrated OpenAI's GPT-5.4 and Codex into its Agent Cloud, enabling enterprises to build and deploy AI agents securely. This partnership aims to streamline AI workflows for real-world applications.
Researchers have extended Courcelle's theorem to show that models of monadic second-order logic (MSO2) formulas with free variables can be represented efficiently using decision diagrams. This advancement could significantly impact parameterized complexity theory and algorithm design.
Researchers developed a model to infer analytical solutions from visualizations of physical fields. This breakthrough could revolutionize AI-assisted scientific discovery.
Researchers introduce AGD-MBRL, a new method that uses advantage estimates to guide diffusion models in reinforcement learning, reducing compounding errors. This approach outperforms traditional policy-only and reward-based guides.
Walnut is a new error tracking tool designed specifically for AI agents, not humans. It's fully compatible with Sentry and operates entirely through a command-line interface.
The race to dominate AI-powered coding tools is intensifying among tech giants. This competition is reshaping how developers work and could redefine the future of software creation.
SuperHQ introduces a new way to run AI coding agents in isolated microVMs, ensuring system safety and clean workflows. The tool provides a unified diff view for managing changes without affecting the host environment.
Researchers introduce Springdrift, a persistent runtime for LLM agents that prioritizes auditing. This could revolutionize how we track and verify AI agent actions.