Tatsu Hashimoto (Stanford University): “Statistical Methods for Trustworthy Language Modeling”
Raisler Lounge (Room 225), Towne Building 220 S 33rd Street, PhiladelphiaPresentation Abstract: Language models work well, but they are far from trustworthy. Major open questions remain on high-stakes issues such as detecting benchmark contamination, identifying LM-generated text, and reliably generating […]