Salvato in:
| Autori principali: | , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2507.00081 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918076656648192 |
|---|---|
| author | Muhoberac, Matthew Parikh, Atharva Vakharia, Nirvi Virani, Saniya Radujevic, Aco Wood, Savannah Verma, Meghav Metaxotos, Dimitri Soundararajan, Jeyaraman Masquelin, Thierry Godfrey, Alexander G. Gardner, Sean Rudnicki, Dobrila Michael, Sam Chopra, Gaurav |
| author_facet | Muhoberac, Matthew Parikh, Atharva Vakharia, Nirvi Virani, Saniya Radujevic, Aco Wood, Savannah Verma, Meghav Metaxotos, Dimitri Soundararajan, Jeyaraman Masquelin, Thierry Godfrey, Alexander G. Gardner, Sean Rudnicki, Dobrila Michael, Sam Chopra, Gaurav |
| contents | Large language models (LLMs) have enabled powerful advances in natural language understanding and generation. Yet their application to complex, real-world scientific workflows remain limited by challenges in memory, planning, and tool integration. Here, we introduce SciBORG (Scientific Bespoke Artificial Intelligence Agents Optimized for Research Goals), a modular agentic framework that allows LLM-based agents to autonomously plan, reason, and achieve robust and reliable domain-specific task execution. Agents are constructed dynamically from source code documentation and augmented with finite-state automata (FSA) memory, enabling persistent state tracking and context-aware decision-making. This approach eliminates the need for manual prompt engineering and allows for robust, scalable deployment across diverse applications via maintaining context across extended workflows and to recover from tool or execution failures. We validate SciBORG through integration with both physical and virtual hardware, such as microwave synthesizers for executing user-specified reactions, with context-aware decision making and demonstrate its use in autonomous multi-step bioassay retrieval from the PubChem database utilizing multi-step planning, reasoning, agent-to-agent communication and coordination for execution of exploratory tasks. Systematic benchmarking shows that SciBORG agents achieve reliable execution, adaptive planning, and interpretable state transitions. Our results show that memory and state awareness are critical enablers of agentic planning and reliability, offering a generalizable foundation for deploying AI agents in complex environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00081 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | State and Memory is All You Need for Robust and Reliable AI Agents Muhoberac, Matthew Parikh, Atharva Vakharia, Nirvi Virani, Saniya Radujevic, Aco Wood, Savannah Verma, Meghav Metaxotos, Dimitri Soundararajan, Jeyaraman Masquelin, Thierry Godfrey, Alexander G. Gardner, Sean Rudnicki, Dobrila Michael, Sam Chopra, Gaurav Multiagent Systems Artificial Intelligence Computation and Language Emerging Technologies Chemical Physics Large language models (LLMs) have enabled powerful advances in natural language understanding and generation. Yet their application to complex, real-world scientific workflows remain limited by challenges in memory, planning, and tool integration. Here, we introduce SciBORG (Scientific Bespoke Artificial Intelligence Agents Optimized for Research Goals), a modular agentic framework that allows LLM-based agents to autonomously plan, reason, and achieve robust and reliable domain-specific task execution. Agents are constructed dynamically from source code documentation and augmented with finite-state automata (FSA) memory, enabling persistent state tracking and context-aware decision-making. This approach eliminates the need for manual prompt engineering and allows for robust, scalable deployment across diverse applications via maintaining context across extended workflows and to recover from tool or execution failures. We validate SciBORG through integration with both physical and virtual hardware, such as microwave synthesizers for executing user-specified reactions, with context-aware decision making and demonstrate its use in autonomous multi-step bioassay retrieval from the PubChem database utilizing multi-step planning, reasoning, agent-to-agent communication and coordination for execution of exploratory tasks. Systematic benchmarking shows that SciBORG agents achieve reliable execution, adaptive planning, and interpretable state transitions. Our results show that memory and state awareness are critical enablers of agentic planning and reliability, offering a generalizable foundation for deploying AI agents in complex environments. |
| title | State and Memory is All You Need for Robust and Reliable AI Agents |
| topic | Multiagent Systems Artificial Intelligence Computation and Language Emerging Technologies Chemical Physics |
| url | https://arxiv.org/abs/2507.00081 |