Showing cs.CLShow all
2 papers · 1 filter
cs.CL2025
Input Matters: Evaluating Input Structure's Impact on LLM Summaries of Sports Play-by-Play
Barkavi Sundararajan, Somayajulu Sripada, Ehud Reiter
A major concern when deploying LLMs in accuracy-critical domains such as sports reporting is that the generated text may not faithfully reflect the input data. We quantify how inpu…
cs.CL2024
Improving Factual Accuracy of Neural Table-to-Text Output by Addressing Input Problems in ToTTo
Barkavi Sundararajan, Somayajulu Sripada, Ehud Reiter
Neural Table-to-Text models tend to hallucinate, producing texts that contain factual errors. We investigate whether such errors in the output can be traced back to problems with t…