Papers by Parthasarathy Ranganathan
Reliable and Cost-Effective Exploratory Data Analysis via Graph-Guided RAG (2025.emnlp-main)
Copied to clipboard
Mossad Helali, Yutai Luo, Tae Jun Ham, Jim Plotts, Ashwin Chaugule, Jichuan Chang, Parthasarathy Ranganathan, Essam Mansour
| Challenge: | Large Language Models (LLMs) have limited accuracy and code reliability on less-studied or private datasets. |
| Approach: | They propose a graph-guided framework that extracts EDA operation semantics from a knowledge graph and then generates executable Python code. |
| Outcome: | Experiments on two datasets show that RAGvis significantly improves code executability, semantic accuracy, and visual quality compared to LLM-only baselines. |