2 papers
cs.CL2026
When RAG Fails to Equalize: Geo-bias in Factual Question Answering over Public Companies
Abhinav Havaldar, Enrico Santus
Retrieval-augmented generation (RAG) is widely assumed to mitigate factual errors in large language models (LLMs), but it remains unclear whether retrieval uniformly compensates fo…
cs.LG2026
GRACE: Gated Refinement for Accurate Causal Edge Discovery in High-Dimensional Time Series
Mohammad Fesanghary, Abhinav Havaldar
From climate teleconnections to gene regulation, modern time-series datasets encompass tens or hundreds of interacting variables, making causal discovery increasingly challenging.…