Tag: Agents

Unlock reliable AI performance! Discover ARC, a new framework tackling 'context rot'-the degradation of information retention in sophisticated agents. Learn ...

Unlock AI's full potential! Learn how researchers are building intelligent, forward-thinking systems-or proactive AI agents-that autonomously handle long-term tasks & ...

Unlock powerful automation! Discover how Amazon leverages strategic AI agent orchestration - fine-tuning workflows for enhanced efficiency & accuracy in ...

Explore why powerful 'agent foresight models' aren't delivering on their promise! New research reveals a surprising disconnect between sophisticated AI ...

Unlock the potential of autonomous systems! Discover how Trajectory Generative Embeddings are revolutionizing & improving imitation learning, enabling robots to ...

Discover AWARE-US, a new benchmark tackling a key challenge in AI: ensuring language models using tool-calling agents truly understand & ...

Discover how AI is evolving beyond chatbots! Explore the exciting rise of  agentic AI, empowering systems to autonomously tackle goals ...

Unlock the next level of AI! Discover MAGMA, a groundbreaking architecture using graph structures to dramatically enhance agent memory. This ...

Explore Alibaba's MAI-UI, a revolutionary framework redefining how we build powerful automated software interfaces. Discover how this new generation of ...

Unlock next-gen AI! CASCADE redefines how artificial intelligence operates, enabling agents to learn, adapt, & evolve through persistent memory & ...

Unlock reliable automation with LLM agents! Agent-C introduces a new framework tackling temporal safety - ensuring tasks are completed correctly ...

Unlock safer, more reliable AI! ROAD offers a fresh approach to tackling the complex issue of agent alignment by automating ...

Unlock enhanced AI capabilities with a new approach! Explore how multi-agent reasoning-deploying multiple specialized agents-overcomes limitations of single LLMs, boosting ...

Unlock a new era of online efficiency! Discover SoDA, a framework enabling seamless & secure agentic web interaction. Reclaim control, ...

Tired of unreliable LLM-powered assistants? Discover Monadic Context Engineering (MCE), a fresh approach for crafting robust AI agent architecture. Learn ...

Explore the crucial practice of AI control monitoring - a vital safeguard for advanced artificial intelligence. Learn how ongoing assessment ...

Discover S$^3$IT, a new Social AI Benchmark designed to rigorously assess embodied agents' ability to navigate complex interactions. This innovative ...

Unlock the next level of AI! Discover PAACE, a groundbreaking framework for agent context engineering that tackles the limitations of ...

Unlock advanced AI capabilities! Discover Turn-PPO, an innovative method simplifying LLM Agent Training for complex tasks like coding and research. ...

Explore how Amazon Bedrock AgentCore Memory transforms conversations into actionable knowledge, enabling AI agents to remember and learn from user ...

New framework COMPASS improves LLM agent reasoning by organizing context and splitting tasks across specialized agents, boosting accuracy up to ...

Unlock peak productivity! Discover how streamlining your tasks with automated processes can revolutionize your work and boost efficiency. Learn practical ...

Learn why customers choose AgentCore to build secure, reliable AI solutions using their choice of frameworks and models for production ...

Unlock the future of automation! Discover how sophisticated software entities, similar to AI agents, are revolutionizing workflows and boosting productivity ...

Explore how decentralized intelligence is reshaping AI! Discover practical applications of collaborative robotics and learn about the potential of a ...

Unlock your AI potential! Discover how to leverage powerful language models for automated grading and personalized feedback—even without coding experience. ...

Explore how AI is evolving beyond simple tasks! This piece dives into the rise of intelligent systems capable of independent ...

ProSEA is a new agent framework that uses exploration & plan evolution for iterative problem-solving. It outperforms baselines on FinanceBench ...

New framework Belief-Calibrated Consensus Seeking (BCCS) improves NLP task accuracy by strategically selecting agent collaborators & calibrating beliefs.

New research reveals a powerful attack (CPS) manipulating web agents by subtly altering images & text, highlighting urgent need for ...