192 stories tagged AI Agents · page 8 of 8
Researchers introduce a novel framework to profile tool-using LLM agents in organizational settings. The A-R space measures action rates and refusal signals, offering insights into agent behavior under different autonomy levels.
IBM Research has introduced VAKRA, a new benchmark for assessing agents' reasoning, tool use, and failure modes. The benchmark aims to provide a standardized way to evaluate agent capabilities across various tasks.
Researchers introduce AlphaEval, a methodology to assess AI agents in real-world scenarios. It addresses gaps in current benchmarks that fail to capture production complexities.
Researchers introduce a new benchmark for evaluating the humanization of mobile GUI agents, framing the challenge as an optimization problem. This work highlights the need for agents to evade detection in human-centric digital ecosystems.
Soul.md is a new open format designed to give AI agents persistent identities. It aims to standardize how AI agents are identified and remembered across different platforms.
Researchers developed proactive AI agents that assist human support teams by learning from unresolved issues and improving continuously. This approach could significantly reduce the workload on human analysts in large-scale cloud service platforms.
Researchers propose a multi-anchor architecture to prevent AI agents from losing their 'self' due to memory overflow. This approach draws inspiration from human memory systems to ensure continuity and resilience.
Researchers introduce SEA-Eval, a benchmark to evaluate self-evolving agents that can learn and adapt across tasks. This addresses limitations of current episodic LLM-based agents.
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.
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.
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.
Researchers introduce Springdrift, a persistent runtime for LLM agents that prioritizes auditing. This could revolutionize how we track and verify AI agent actions.
A new open-source project, Entroly, introduces a self-evolving daemon that analyzes and improves codebases autonomously. It promises to revolutionize AI agent development by continuously optimizing code.
Mugib introduces AI agents capable of operating across chat, voice, web, and live data channels. This innovation promises to revolutionize customer interactions by providing a unified experience.
Sierra has launched Ghostwriter, an agent-building agent that replaces traditional web applications with natural language interactions. This shift could redefine how users interact with software.
A new study highlights the risks of malicious attacks on the LLM supply chain, demonstrating how agents can be compromised. The findings underscore the need for robust security measures in AI development.
Qualixar OS emerges as the first application-layer operating system designed for universal AI agent orchestration, supporting 10 LLM providers and 8+ frameworks. It introduces execution semantics for 12 distinct multi-agent topologies and a novel LLM-driven design engine called Forge.
Poke introduces AI agents accessible via text message, eliminating the need for complex apps or technical expertise. This democratizes AI automation for everyday users.
Researchers introduce IntentScore, a new reward model designed to evaluate the quality of actions taken by Computer-Use Agents (CUAs) to prevent irreversible errors. Trained on 398,000 offline GUI interaction steps across three operating systems, it uses contrastive alignment and margin ranking to ensure actions align with user intent.
Anthropic has launched Cowork, a new agent for Claude Desktop that automates file tasks without coding, built in just 10 days using Claude Code itself. This move positions Anthropic to directly challenge Microsoft Copilot in the AI productivity market.
IBM Research has released ALTK-Evolve, an open-source framework enabling AI agents to learn and adapt in real-time during task execution. This approach eliminates the need for static pre-training by allowing agents to evolve their strategies based on immediate feedback.
QVeris introduces AI agents that can discover, inspect, and call 10,000 capabilities via one protocol. This innovation simplifies interactions with various capabilities.
CongaLine is a self-hosted AI agent fleet that allows teams and individuals to run isolated AI assistants. It prioritizes security and offers flexible deployment options.