This paper proposes a query-level meta-agent named FlowReasoner to automate the design of query-level multi-agent systems, i.e., one system per user query. Our core idea is to incentivize a reasoning-based meta-agent via external execution feedback. Concretely, by distilling DeepSeek R1, we first endow the basic reasoning ability regarding the generation of multi-agent systems to FlowReasoner. Then, we further enhance it via reinforcement learning (RL) with external execution feedback. A multi-purpose reward is designed to guide the RL training from aspects of performance, complexity, and efficiency. In this manner, FlowReasoner is enabled to generate a personalized multi-agent system for each user query via deliberative reasoning. Experiments on both engineering and competition code benchmarks demonstrate the superiority of FlowReasoner. Remarkably, it surpasses o1-mini by 10.52% accuracy across three benchmarks. The code is available at https://github.com/sail-sg/FlowReasoner.

The emergence of Large Language Models (LLMs) in Multi-Agent Systems (MAS) has opened new possibilities for artificial intelligence, yet current implementations face significant challenges in resource management, task coordination, and system efficiency. While existing frameworks demonstrate the potential of LLM-based agents in collaborative problem-solving, they often lack sophisticated mechanisms for parallel execution and dynamic task management. This paper introduces DynTaskMAS, a novel framework that orchestrates asynchronous and parallel operations in LLM-based MAS through dynamic task graphs. The framework features four key innovations: (1) a Dynamic Task Graph Generator that intelligently decomposes complex tasks while maintaining logical dependencies, (2) an Asynchronous Parallel Execution Engine that optimizes resource utilization through efficient task scheduling, (3) a Semantic-Aware Context Management System that enables efficient information sharing among agents, and (4) an Adaptive Workflow Manager that dynamically optimizes system performance. Experimental evaluations demonstrate that DynTaskMAS achieves significant improvements over traditional approaches: a 21-33% reduction in execution time across task complexities (with higher gains for more complex tasks), a 35.4% improvement in resource utilization (from 65% to 88%), and near-linear throughput scaling up to 16 concurrent agents (3.47X improvement for 4X agents). Our framework establishes a foundation for building scalable, high-performance LLM-based multi-agent systems capable of handling complex, dynamic tasks efficiently.

Understanding the meaning of data is crucial for performing data analysis, yet for the users to gain insight into the content and structure of their database, a tedious data exploration process is often required [2, 16]. A common industry practice taken on by specialists such as Knowledge Engineers is to explicitly construct an intermediate layer between the database and the user -- a semantic layer -- abstracting away certain details of the database schema in favor of clearer data semantics [3, 10]. In the era of Large Language Models (LLMs), industry practitioners and researchers attempt to circumvent this costly process using LLM-powered Natural Language Interfaces [4, 6, 12, 18, 19, 22]. The promise of such Text-to-SQL solutions is to allow users without technical expertise to seamlessly interact with databases. For example, a new company employee could effectively issue queries in natural language without programming expertise or even explicit knowledge of the database structure, e.g., knowing the names of entities or properties, the exact location of data sources, etc.

This paper investigates the interactions between multiple agents within Large Language Models (LLMs) in the context of programming and coding tasks. We utilize the AutoGen framework to facilitate communication among agents, evaluating different configurations based on the success rates from 40 random runs for each setup. The study focuses on developing a flexible automation framework for applying the Finite Element Method (FEM) to solve linear elastic problems. Our findings emphasize the importance of optimizing agent roles and clearly defining their responsibilities, rather than merely increasing the number of agents. Effective collaboration among agents is shown to be crucial for addressing general FEM challenges. This research demonstrates the potential of LLM multi-agent systems to enhance computational automation in simulation methodologies, paving the way for future advancements in engineering and artificial intelligence.

Due to the growing complexity of modern Integrated Circuits (ICs), automating hardware design can prevent a significant amount of human error from the engineering process and result in less errors. Verilog is a popular hardware description language for designing and modeling digital systems; thus, Verilog generation is one of the emerging areas of research to facilitate the design process. In this work, we propose VerilogCoder, a system of multiple Artificial Intelligence (AI) agents for Verilog code generation, to autonomously write Verilog code and fix syntax and functional errors using collaborative Verilog tools (i.e., syntax checker, simulator, and waveform tracer). Firstly, we propose a task planner that utilizes a novel Task and Circuit Relation Graph retrieval method to construct a holistic plan based on module descriptions. To debug and fix functional errors, we develop a novel and efficient abstract syntax tree (AST)-based waveform tracing tool, which is integrated within the autonomous Verilog completion flow. The proposed methodology successfully generates 94.2% syntactically and functionally correct Verilog code, surpassing the state-of-the-art methods by 33.9% on the VerilogEval-Human v2 benchmark.

Large Language Models (LLMs) have demonstrated significant potential in causal discovery tasks by utilizing their vast expert knowledge from extensive text corpora. However, the multi-agent capabilities of LLMs in causal discovery remain underexplored. This paper introduces a general framework to investigate this potential. The first is the Meta Agents Model, which relies exclusively on reasoning and discussions among LLM agents to conduct causal discovery. The second is the Coding Agents Model, which leverages the agents' ability to plan, write, and execute code, utilizing advanced statistical libraries for causal discovery. The third is the Hybrid Model, which integrates both the Meta Agents Model and Coding Agents Model approaches, combining the statistical analysis and reasoning skills of multiple agents. Our proposed framework shows promising results by effectively utilizing LLMs' expert knowledge, reasoning capabilities, multi-agent cooperation, and statistical causal methods. By exploring the multi-agent potential of LLMs, we aim to establish a foundation for further research in utilizing LLMs multi-agent for solving causal-related problems.

The global medical industry experienced a 13% rise in new job postings related to artificial intelligence in Q3 2022 compared with the previous quarter, according to GlobalData's Jobs Analytics. This compares to a 13% increase in the previous quarter and a 39% increase versus Q3 2021. Notably, Software and Web Developers, Programmers, and Testers jobs accounted for a 16% share of the global medical industry's artificial intelligence-related total new job postings in Q3 2022, down 11% over the prior quarter. Software and Web Developers, Programmers, and Testers, with a share of 16%, emerged as the top artificial intelligence-related job roles within the medical industry in Q3 2022, with new job postings drop by 11% quarter-on-quarter. General and Operations Managers came in second with a share of 5% in Q3 2022, with new job postings rise by 4% over the previous quarter.

According to the National Interagency Fire Center, wildfires have burned 2,990,255 acres this year (June 17). Adding to firefighters' challenges, using current resources, it often takes hours to map a growing wildfire's perimeter and heat spots. It sometimes takes days using fuel property data – often 3-5 years old – to help predict fire behavior. Time is not on their side and the situation on the ground is always changing. Lockheed Martin is using AI/ML to help get critical data to firefighters faster.

Petuum, the creator of the world's first composable platform for MLOps, and the Inception Institute for Artificial Intelligence (IIAI), have agreed to partner on the development of revolutionary AI applications. Petuum has recently announced a limited release of the composable platform, which includes the AI OS, Universal Pipelines, Deployment Manager, and Experiment Manager, for select private beta partners. Through the partnership with Petuum, IIAI's enterprise AI/ML teams will operationalize and scale their applications into production. Founded in 2018, IIAI's mission is to build full-stack AI solutions and operating systems for enterprise businesses and developers. Besides being the research arm for G42, IIAI is also empowering stakeholders with AI applications and incubating new technology at the cutting edge of ML innovation.

Nine months since the launch of our first commercial product, the OpenAI API, more than 300 applications are now using GPT-3, and tens of thousands of developers around the globe are building on our platform. They also want a way to edit their address in checkout and save multiple payment methods." "GPT-3's ability to identify themes from natural language and generate summaries allows Viable to give product, customer experience, and marketing teams at companies across industries a better understanding of their customers' wants and needs," said Daniel Erickson, CEO of Viable. Algolia uses GPT-3 in their Algolia Answers product to offer relevant, lightning-fast semantic search for their customers. When the OpenAI API launched, Algolia partnered with OpenAI to integrate GPT-3 with their advanced search technology in order to create their new Answers product that better understands customers' questions and connects them to the specific part of the content that answers their questions. Algolia Answers helps publishers and customer support help desks query in natural language and surface nontrivial answers. "We've seen great results from Algolia Answers on questions that are difficult to answer with textual search alone," said Peter Buffington, Product Manager at ABC Australia. "It was able to return very relevant, evergreen content from our news archives for questions such as'Why does a volcano erupt?'" "GPT-3 allows Algolia to answer more complex queries than ever before with our Algolia Answers product, identifying deeper contextual information to improve the quality of results and deliver them in seconds," said Dustin Coates, Product and GTM Manager at Algolia. We require developers to implement safety measures such as rate limits, user verification and testing, or human-in-the-loop requirements before they move into production. Nine months since the launch of our first commercial product, the OpenAI API, more than 300 applications are now using GPT-3, and tens of thousands of developers around the globe are building on our platform. They also want a way to edit their address in checkout and save multiple payment methods."