Tag
发现一篇关于Agentic AI的全面指南，从基础到系统架构，覆盖了智能体AI的全貌。
Discusses multi-agent systems designed to handle complex tasks, likely covering coordination and collaboration among AI agents.
The article proposes that AI agent recommendation systems should incorporate a merchant feedback mechanism to improve their effectiveness and reliability.
A senior Google engineer released a free 421-page document covering agentic design patterns for AI systems, with code-backed chapters on prompt chaining, multi-agent coordination, guardrails, and reasoning.
A step-by-step guide to building AI systems using LLMs, covering steps from choosing models to evaluation using frameworks, vector databases, and data extraction tools.
This paper proposes a BFT-derived protocol for epistemic synthesis enabling emergent collaborative deliberation among multiple AI models.
This article explores how concepts from urban economics, such as traffic, zoning, and pollution, can model externalities in agentic AI systems. It introduces a Behavioral Externality Multiplier (BEM) and proposes a layered framework involving architecture, substrate, and governance to measure and mitigate costly consequences of cheap AI actions.
A tweet by Martin Casado highlighting a solution to the difficult problem of exposing traces at scale to AI agents, balancing cost and AI leverage.
Prompt Logic Gates (PLG) is a visual prompt engineering experiment that organizes prompts using semantic logic gates (AND, OR, NOT, Ask Questions) to manage complex system-like prompts, aiming to improve maintainability and consistency.
This article argues that while AI excels at pattern recognition and hypothesis generation, scientific and economic progress requires grounded interaction with reality and institutional execution, emphasizing the need for human-AI collaboration.
The article argues that the next major AI shift will be towards systems with reliable operational memory—able to remember, update, and use knowledge over time—rather than just building smarter models.
This paper introduces the concepts of Agentic Technical Debt and Stochastic Tax, defining new liabilities and operating costs specific to agentic AI systems that combine stochastic models with tool use and workflows, and proposes lightweight governance controls.
The article highlights the problem of AI memory becoming unreliable after six months, with contradictions and drifted summaries, and questions whether the industry is focusing on adding more storage rather than improving maintainability.
This article discusses the challenges of operational drift in deployed AI systems, questioning whether model quality, data, or business processes break first after deployment.
This paper introduces Computable Fair Division (CFD), a framework using Boltzmann-Softmax control to balance efficiency and fairness in AI resource allocation, with real-time adaptation via AHC++.
An op-ed discussing the gap between AI code generation and production-grade systems, emphasizing that human judgment and domain expertise remain critical for orchestrating interconnected decision loops in complex domains.
A tweet highlights a 40-minute masterclass by the founder of a $20B Chinese AI company, explaining Agent Swarms and AI systems at scale, with the implication that the architecture beats Anthropic's Claude.
A survey paper examining the transition of AI from task-specific assistants to workflow-level research automators, defining AutoResearch as the spectrum of AI-powered scientific workflow automation and analyzing challenges in autonomy, reproducibility, and accountability.
Mark Saroufim gave a keynote at the MLSys conference covering the evolution of AI systems, why AI is needed to improve them, and promising future directions. The recording will be released soon.
The IMF published a formal note introducing the concept of agentic payments, framing the tension between probabilistic AI systems and deterministic payment infrastructure, and defining the shift from 'click to pay' to 'decide to pay'.