arXiv cs.AIWednesday · May 27, 2026FREE

Experiments in Agentic AI for Science

agentic-aiscientific-workflowsllmragarxiv

This arXiv paper (2605.26305) introduces two agentic AI frameworks for scientific workflows. DeepTS/DeepCollector automates large-scale curation, extraction, and deduplication of time-series datasets. DeepScribe is an autonomous presentation analyzer that converts visually dense physics lectures into structured scientific reports. Both use a hybrid Local Body, Remote Brain architecture via Google Colab, with Python orchestrators invoking LLM cloud backends. Key engineering contributions include granular attribute extraction (Cellular RAG), remote data inspection, and distributed concurrency controls to overcome LLM context and reasoning limits. The authors generalize DeepTS to support deep knowledge graphs and discuss application to high-energy physics (DeepQCD). The paper is available on arXiv, published May 27, 2026.

// why it matters

Enables autonomous AI agents to handle complex scientific data curation and analysis tasks.

Sources

Primary · arXiv cs.AI

▸ Read original at arxiv.org

Cross-Entropy Games and Frost Training Hierarchical Prompt-Domain Control and Learning for Resource-Constrained Agentic Language Models A Fixed-Budget, Cluster-Aware Standard for LLM-as-a-Judge Evaluation: A Multi-Hop RAG Stress Test

Experiments in Agentic AI for Science

Sources

Related

Like this? Get the next digest.