[PDG 474] AFlow: Automating Agentic Workflow Generation

Link to article: https://arxiv.org/pdf/2410.10762
Title: AFlow: Automating Agentic Workflow Generation (ICLR 2025):
Track: Automated Agent Optimization
Content: AFLOW is an automated framework that uses Monte Carlo Tree Search to optimize LLM agentic workflows represented as code, eliminating the need for manual workflow construction. It achieves a 5.7% improvement over existing methods and enables smaller models to outperform GPT-4o on certain tasks at a fraction of the cost.
Slack link: ml-ka.slack.com, channel: #pdg. Please join us -- if you cannot join, please message us here or to mlpaperdiscussiongroupka@gmail.com.

In the Paper Discussion Group (PDG) we discuss recent and fundamental papers in the area of machine learning on a weekly basis. If you are interested, please read the paper beforehand and join us for the discussion. If you have not fully understood the paper, you can still participate – everyone is welcome! You can join the discussion or simply listen in. The discussion is in German or English depending on the participants.