Paper 2604.27351
Heterogeneous Scientific Foundation Model Collaboration
- Published
- Apr 2026
- Research lab
- Independent
- Citations
- 1
- GitHub
- 23 stars
01 In brief
Summary
The paper introduces Eywa, a heterogeneous agentic framework that integrates domain-specific foundation models (FMs) with large language model (LLM) agents to solve scientific tasks involving non-linguistic data like time series and tabular data.
Eywa uses an FM-LLM 'Tsaheylu' interface, implemented via the Model Context Protocol, allowing LLMs to configure and interpret FM outputs.
Three instantiations are proposed: EywaAgent (single-agent), EywaMAS (multi-agent), and EywaOrchestra (dynamic planning-based orchestration).
The authors introduce EywaBench, a benchmark spanning physical, life, and social sciences.
Experiments show Eywa improves utility by ~7% over single-LLM baselines while reducing token usage by ~30% and execution time by ~10%, with similar gains in multi-agent settings.
Theoretical analysis proves strict risk improvements under a domain advantage assumption.
The framework is robust across LLM backbones and hyperparameters, and EywaOrchestra achieves near-expert performance with lower cost and automation.
The work addresses the limitation of language-centric systems in scientific domains by enabling modality-native collaboration with specialized FMs.
02 From the paper
Abstract
Agentic large language model systems have demonstrated strong capabilities. However, their reliance on language as the universal interface fundamentally limits their applicability to many real-world problems, especially in scientific domains where domain-specific foundation models have been developed to address specialized tasks beyond natural language. In this work, we introduce Eywa, a heterogeneous agentic framework designed to extend language-centric systems to a broader class of scientific foundation models. The key idea of Eywa is to augment domain-specific foundation models with a language-model-based reasoning interface, enabling language models to guide inference over non-linguistic data modalities. This design allows predictive foundation models, which are typically optimized for specialized data and tasks, to participate in higher-level reasoning and decision-making processes within agentic systems. Eywa can serve as a drop-in replacement for a single-agent pipeline (EywaAgent) or be integrated into existing multi-agent systems by replacing traditional agents with specialized agents (EywaMAS). We further investigate a planning-based orchestration framework in which a planner dynamically coordinates traditional agents and Eywa agents to solve complex tasks across heterogeneous data modalities (EywaOrchestra). We evaluate Eywa across a diverse set of scientific domains spanning physical, life, and social sciences. Experimental results demonstrate that Eywa improves performance on tasks involving structured and domain-specific data, while reducing reliance on language-based reasoning through effective collaboration with specialized foundation models.