cs.AINov 15, 2026

Towards Agentic Scientific Discovery in Distributed Systems

Authors: Rashmita Soni, Abhirup Ghosh, Soumya Banerjee

Organizations: Indian Institute of Technology Delhi, India · University of Birmingham, UK · University of Cambridge, UK

Abstract

We investigate when adaptive tool selection helps identify a shared equation across clients with different input coverage. Query-ATLAS (query-conditioned Agentic Tool-based Learning for Aggregating Symbolic Equations) uses an LLM to guide symbolic search from client errors and previous proposals in a simulated distributed system. A four-architecture comparison reveals task-dependent opportunities: relative to fixed PySR-LLM, validation acceptances rise from 3/12 to 7/12 on CRK3 and from 11/12 to 12/12 on oscillator2, while MatSci3 favors the fixed workflow. Across 72 settings in a post-hoc six-task subset of eight development tasks, Query-ATLAS achieves 64 validation and 64 in-distribution passes, versus 62 and 59 for the fixed workflow; both pass 47/60 eligible out-of-distribution tests. The fixed workflow retains lower median errors and cost. Broader query coverage emerges as a promising benefit, motivating task-aware tool selection and client-aware fitting for scientific agents.

Figures & tables

Explore similar work

CardsList
  1. Coupled Harness--Model Evolution for Scientific Equation Discovery

    Oct 13, 2026Pengze Li, Jiaquan Zhang, Encheng Su +8Large Language Model-Guided Scientific DiscoveryScientific Discovery

  2. El Agente Forjador: Task-Driven Agent Generation for Quantum Simulation

    Apr 16, 2026Zijian Zhang, Aiwei Yin, Amaan Baweja +4Multi-Agent LLM SystemsAI for Science

  3. SciToolAgent-Evo: An Ontology-Aware Self-Evolving Agent for Open-World Scientific Tool Acquisition

    Jul 30, 2026Yuqi Tang, Chenyi Zhou, Libin Wang +3AI for ScienceLLM Agent Skill Learning