Process-Reward Tactic Evolution for Long-Horizon Bioinformatics Workflows

2026-06-18 · Lingzhi Yang, Yubo Fan, Song Wu, Gilchan Park arxiv

LLM agents can write code and call tools, but reliable bioinformatics work requires long-horizon interaction with workflow software, typed data objects, provenance, and biological checks. We study this setting through Galaxy workflow execution. The agent must explore task data, construct or adapt an executable workflow DAG, bind inputs and dataset collections, monitor execution, debug failures, and validate biological outputs. We propose Process-Reward Tactic Evolution, a Galaxy-based training framework that turns verified workflow rollouts into reusable \tactics. During training, agents practice on curriculum-organized Galaxy tasks in Agent Gym; process verifiers score workflow construction, software interaction, execution, and biological correctness; successful and failed traces are distilled into a tactic library. At inference, the trained executor, Process-Reward Tactic Evolution, uses this library to execute held-out peer reviewed Galaxy workflow converted BioWorkflow Bench and BioAgent Bench tasks in isolated environments. The paper evaluates whether process-supervised tactic accumulation improves long-horizon bioinformatics workflow completion, biological correctness, and execution efficiency over no-memory and reflection-style baselines.

📄 PDF Abstract BibTeX arXiv:2606.20839

Code (0)

등록된 구현이 없습니다.

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

2026-08-18 · Zhi Zheng, Rongsheng Chen, Yunpeng Ba, Zhenkun Wang 외 arxiv

Reinforcement Learning (RL) has been promising in single-turn LLM fine-tuning. However, long-horizon agentic reasoning introduces increasingly branching interactions and sparse rewards, exposing several limitations of RL…

Reinforcement Learning

Evolution Guided Generative Flow Networks

2024-02-03 · Zarif Ikram, Ling Pan, Dianbo Liu

Generative Flow Networks (GFlowNets) are a family of probabilistic generative models that learn to sample compositional objects proportional to their rewards. One big challenge of GFlowNets is training them effectively w…

Evolutionary Algorithms

CODE-SHARP: Continuous Open-ended Discovery and Evolution of Skills as Hierarchical Reward Programs

2026-02-10 · Richard Bornemann, Pierluigi Vito Amadori, Antoine Cully arxiv

A core quality of general intelligence is the ability to open-endedly expand and evolve its set of mastered skills autonomously. While recent Foundation Model (FM) driven approaches have shown promising results towards t…

Reinforcement Learning

MagicWorld: Towards Long-Horizon Stability for Interactive Video World Exploration

2025-11-24 · Guangyuan Li, Bo Li, Jinwei Chen, Xiaobin Hu 외 arxiv

Recent interactive video world model methods generate scene evolution conditioned on user instructions. Although they achieve impressive results, two key limitations remain. First, they exhibit motion drift in complex en…

DemoEvolve: Overcoming Sparse Feedback in Agentic Harness Evolution with Demonstrations

2026-05-23 · Lirong Che, Yuzhe yang, Peiwen lin, Chuang wang 외 arxiv

Agent harness evolution improves frozen language-model agents by modifying the executable structures around them. We study this paradigm as a form of sample-efficient fast adaptation: instead of updating model weights, a…