Episodic Memory in Agentic Frameworks: Suggesting Next Tasks
This addresses the problem of improving workflow creation in human-AI co-creation for scientific applications, but it appears incremental as it builds on existing agentic frameworks with a memory-based enhancement.
The paper tackles the challenge of recommending next steps in agentic frameworks for scientific workflows by proposing an episodic memory architecture that stores and retrieves past workflows to guide task suggestions, avoiding reliance on LLMs that risk hallucination and require fine-tuning with scarce data.
Agentic frameworks powered by Large Language Models (LLMs) can be useful tools in scientific workflows by enabling human-AI co-creation. A key challenge is recommending the next steps during workflow creation without relying solely on LLMs, which risk hallucination and require fine-tuning with scarce proprietary data. We propose an episodic memory architecture that stores and retrieves past workflows to guide agents in suggesting plausible next tasks. By matching current workflows with historical sequences, agents can recommend steps based on prior patterns.