Semantic Retrieval
Papers
No papers found.
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Human baseline | Soft-F1 | 0.84 | — | Unverified |
| 2 | k-NN with sentence n-grams, GPT-2 embeddings, fICA | Soft-F1 | 0.51 | — | Unverified |
| 3 | DBTW, GPT-1 embeddings, fICA | Soft-F1 | 0.51 | — | Unverified |
| 4 | LSA baseline | Soft-F1 | 0.39 | — | Unverified |
| 5 | Universal Sentence Encoder | Soft-F1 | 0.38 | — | Unverified |
| 6 | Sentence BERT | Soft-F1 | 0.31 | — | Unverified |