A self-improving LLM agentic framework that generates, validates, and caches VTK visualization scripts, uses a fine-tuned vision model to answer feature queries, and demonstrates improved isovalue selection on four volumetric datasets.
FREYR: A Framework for Recognizing and Executing Your Requests
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Large language models excel as conversational agents, but their capabilities can be further extended through tool usage, i.e.: executable code, to enhance response accuracy or address specialized domains. Current approaches to enable tool usage often rely on model-specific prompting or fine-tuning a model for function-calling instructions. Both approaches have notable limitations, including reduced adaptability to unseen tools and high resource requirements. This paper introduces FREYR, a streamlined framework that modularizes the tool usage process into separate steps. Through this decomposition, we show that FREYR achieves superior performance compared to conventional tool usage methods. We evaluate FREYR on a set of real-world test cases specific for video game design and compare it against traditional tool usage as provided by the Ollama API.
citation-role summary
background 1
citation-polarity summary
fields
cs.HC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
VizGenie: Toward Self-Refining, Domain-Aware Workflows for Next-Generation Scientific Visualization
A self-improving LLM agentic framework that generates, validates, and caches VTK visualization scripts, uses a fine-tuned vision model to answer feature queries, and demonstrates improved isovalue selection on four volumetric datasets.