Few-Shot Text Classification
Few-shot Text Classification predicts the semantic label of a given text with a handful of supporting instances 1
Papers
No papers found.
All datasetsRAFTAverage on NLP datasetsAmazon CounterfeitODIC 10-way (10-shot)ODIC 10-way (5-shot)ODIC 5-way (10-shot)ODIC 5-way (5-shot)SST-5
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | T-Few | Avg | 0.76 | — | Unverified |
| 2 | Human (crowdsourced) | Avg | 0.74 | — | Unverified |
| 3 | GPT-3 | Avg | 0.63 | — | Unverified |
| 4 | AdaBoost | Avg | 0.51 | — | Unverified |
| 5 | GPT-Neo | Avg | 0.48 | — | Unverified |
| 6 | GPT-2 | Avg | 0.46 | — | Unverified |
| 7 | BART MNLI zero-shot | Avg | 0.38 | — | Unverified |
| 8 | Plurality-class | Avg | 0.33 | — | Unverified |
| 9 | GPT-3 zero-shot | Avg | 0.29 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SetFit + OCD(5) | Accuracy | 0.65 | — | Unverified |
| 2 | SetFit + OCD | Accuracy | 0.64 | — | Unverified |
| 3 | T-few 3B | Accuracy | 0.63 | — | Unverified |
| 4 | SetFit | Accuracy | 0.62 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SetFit + OCD | Accuracy | 0.41 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Induction Networks | Accuracy | 81.64 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Induction Networks | Accuracy | 78.27 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Induction Networks | Accuracy | 88.49 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Induction Networks | Accuracy | 87.16 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | SetFit + OCD | Accuracy | 0.48 | — | Unverified |