TED-talks
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Likes Prediction | TED Talks | Beta (β)0.339 | 21 | |
| Regression predicting Views | TED talks dataset | Beta Coefficient0.314 | 21 | |
| Drag-style image editing | Ted-talks (test) | FID51.51 | 9 | |
| Human Image Animation | TED-talks (test) | AKD2.16 | 8 | |
| Video Reenactment | TED-talks dataset (test) | IE0.93 | 7 | |
| 2D Character Animation | TED-talks dataset | FVD129.66 | 6 | |
| Code-switching mitigation | TED Talks English-Hindi Llama-3.2-1B n=500 (test) | CSI81 | 3 | |
| Code-switching mitigation | TED Talks English-Russian Llama-3.2-1B, n=500 (test) | CSI1 | 3 | |
| Code-switching mitigation | TED Talks English-Spanish Llama-3.2-1B, n=500 (test) | CSI62 | 3 | |
| Code-switching mitigation | TED Talks English-Chinese Llama-3.2-1B, n=500 (test) | CSI64 | 3 | |
| Machine Translation | TED Talks Arabic-English ar-en (test) | BLEU32.6 | 3 | |
| Machine Translation | TED Talks Hebrew-English (he-en) (test) | BLEU37.3 | 3 | |
| Machine Translation | TED Talks Italian-English it-en (test) | BLEU39.8 | 3 | |
| Machine Translation | TED Talks German-English (de-en) (test) | BLEU37.5 | 3 | |
| Video reconstruction | TED-talks | L1 Loss0.026 | 3 | |
| Video reconstruction | TED-talks (test) | L1 Loss0.026 | 2 | |
| Automatic Speech Recognition | TED-talks 2012 (test) | WER14.4 | 2 | |
| Automatic Speech Recognition | TED-talks 2011 (TST) | WER13.8 | 2 | |
| Image Animation | TED-talks (test) | Continuity63.93 | 1 | |
| Animation Quality | TED-talks (user study) | Preference vs FOMM (%)96.6 | 1 |
Showing 20 of 20 rows