232 stories tagged AI Agents · page 9 of 10
A new study highlights how AI agents can be misled by adversarial environments that manipulate tool outputs. The research introduces the concept of Adversarial Environmental Injection (AEI) to formalize this security risk.
Researchers introduce AutomationBench, a new benchmark for evaluating AI agents on complex, cross-application workflows. It tests coordination, API discovery, and policy adherence in real-world business scenarios.
NeoCognition, founded by an OSU researcher, has raised $40M to build AI agents capable of mastering any domain through human-like learning. The funding highlights the growing interest in advanced AI learning mechanisms.
Researchers propose PolicyBank, a framework where LLM agents refine their policy interpretations through interaction and feedback. This could reduce gaps between natural language specifications and actual agent behavior.
Researchers propose a framework to unify memory, skills, and rules in LLM agents, addressing the challenge of managing accumulated experience. The study highlights a lack of cross-community collaboration in the field.
DeepMind has outlined key pitfalls in AI agent development, highlighting risks like goal misalignment and reward hacking. The findings stress the need for robust safety measures in autonomous systems.
Web Agent Bridge introduces an open-source operating system for AI agents, combining MIT and Open Core licensing. It aims to standardize AI agent development and deployment.
A new open-source project called Rigor aims to prevent AI services from degrading over time. It acts as a proxy for OpenCode and Claude Code, grounding projects in an epistemic graph with LLM-based evaluation.
Startups are grappling with how to safely provide AI/ML teams with production database access. The challenge differs from traditional BI tools in ways that aren't yet fully understood. The question of where agents should connect—primary, read replica, or warehouse—remains unresolved.
Operating 14 AI agents for half a year revealed critical insights on scalability, cost, and reliability. The experience highlights the need for robust infrastructure and continuous monitoring.
Researchers propose NuHF Claw, a risk-constrained AI agent framework designed to support human operators in digital nuclear control rooms. The system aims to mitigate cognitive risks while maintaining human authority in safety-critical environments.
Researchers introduce Credo, a framework for declarative control of LLM pipelines using beliefs and policies. It aims to make agent behavior more transparent and adaptable than current imperative approaches.
Fleeks is a new infrastructure platform designed to remove bottlenecks for AI agents, enabling them to execute, verify, and integrate code seamlessly. The platform aims to bridge the gap between code generation and real-world application.
Researchers introduce a lightweight parallel monitoring system to detect and recover from reasoning degradation in LLM agents. The Cognitive Companion reduces overhead compared to existing solutions. The study shows promise in mitigating issues like looping and drift in multi-step tasks.
OpenAI has updated its Agents SDK with native sandbox execution and a model-native harness, enabling developers to build secure, long-running agents. This advancement simplifies the integration of AI agents across various files and tools.
Researchers introduce HORIZON, a diagnostic benchmark to analyze failures in LLM-based agents on long-horizon tasks. The study highlights the need for better characterization of these failures to improve agentic systems.
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.