paper-with-me

Papers

FlowReasoner: Reinforcing Query-Level Meta-Agents

2025-04-21 · Hongcheng Gao, Yue Liu, Yufei He, Longxu Dou, Chao Du, Zhijie Deng, Bryan Hooi, Min Lin, Tianyu Pang

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.

📄 PDF Abstract BibTeX arXiv:2504.15257

Code (1)

sail-sg/flowreasoner 공식 구현

Tasks

Reinforcement Learning (RL)

Similar Papers 제목 키워드 기반

MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild

2026-03-17 · Peng Xia, Jianwen Chen, Xinyu Yang, Haoqin Tu 외 arxiv

Large language model (LLM) agents are increasingly used for complex tasks, yet deployed agents often remain static, failing to adapt as user needs evolve. This creates a tension between the need for continuous service an…

Reinforcement Learning

EvoGround: Self-Evolving Video Agents for Video Temporal Grounding

2026-05-13 · Minjoon Jung, Byoung-Tak Zhang, Lorenzo Torresani arxiv

Video temporal grounding (VTG) takes an untrimmed video and a natural-language query as input and localizes the temporal moment that best matches the query. Existing methods rely on large, task-specific datasets requirin…

Galaxy: A Cognition-Centered Framework for Proactive, Privacy-Preserving, and Self-Evolving LLM Agents

2025-08-06 · Chongyu Bao, Ruimin Dai, Yangbo Shen, Runyang Jian 외 arxiv

Intelligent personal assistants (IPAs) such as Siri and Google Assistant are designed to enhance human capabilities and perform tasks on behalf of users. The emergence of LLM agents brings new opportunities for the devel…

Learning to Coordinate Multiple Reinforcement Learning Agents for Diverse Query Reformulation

2018-09-27 · ICLR 2019 5 · Rodrigo Nogueira, Jannis Bulian, Massimiliano Ciaramita

We propose a method to efficiently learn diverse strategies in reinforcement learning for query reformulation in the tasks of document retrieval and question answering. In the proposed framework an agent consists of mult…

DiversityQuestion Answeringreinforcement-learningReinforcement Learning+2

Reinforcing Generated Images via Meta-learning for One-Shot Fine-Grained Visual Recognition

2022-04-22 · Satoshi Tsutsui, Yanwei Fu, David Crandall

One-shot fine-grained visual recognition often suffers from the problem of having few training examples for new fine-grained classes. To alleviate this problem, off-the-shelf image generation techniques based on Generati…

DiversityFine-Grained Image ClassificationFine-Grained Visual Recognitionimage-classification+4