Institutsseminar/2021-07-09
Zur Navigation springen
Zur Suche springen
Datum | Fr 9. Juli 2021, 14:00 Uhr | |
---|---|---|
Dauer | 30 min | |
Ort | ||
Webkonferenz | https://sdqweb.ipd.kit.edu/wiki/Institutsseminar/Microsoft Teams | |
Vorheriger Termin | Fr 2. Juli 2021 | |
Nächster Termin | Fr 16. Juli 2021 |
Vorträge
Vortragende(r) | Dennis Grötzinger | |
---|---|---|
Titel | Exploring The Robustness Of The Natural Language Inference Capabilties Of T5 | |
Vortragstyp | Bachelorarbeit | |
Betreuer(in) | Jan Keim | |
Vortragsmodus | ||
Kurzfassung | Large language models like T5 perform excellently on various NLI benchmarks. However, it has been shown that even small changes in the structure of these tasks can significantly reduce accuracy. I build upon this insight and explore how robust the NLI skills of T5 are in three scenarios. First, I show that T5 is robust to some variations in the MNLI pattern, while others degenerate performance significantly. Second, I observe that some other patterns that T5 was trained on can be substituted for the MNLI pattern and still achieve good results. Third, I demonstrate that the MNLI pattern translate well to other NLI datasets, even improving accuracy by 13% in the case of RTE. All things considered, I conclude that the robustness of the NLI skills of T5 really depend on which alterations are applied. |
- Neuen Vortrag erstellen