Chunking
Chunking, also known as shallow parsing, identifies continuous spans of tokens that form syntactic units such as noun phrases or verb phrases.
Example:
| Vinken | , | 61 | years | old | | --- | ---| --- | --- | --- | | B-NLP| I-NP | I-NP | I-NP | I-NP |
Papers
Showing 1–10 of 447 papers
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ACE | Exact Span F1 | 97.3 | — | Unverified |
| 2 | BERT-CRF (Replicated in AdaSeq) | Exact Span F1 | 97.18 | — | Unverified |
| 3 | ELMo + MAT + Multi-Task | Exact Span F1 | 97.04 | — | Unverified |
| 4 | CVT+Multi-Task+Large | Exact Span F1 | 96.98 | — | Unverified |
| 5 | ELMo + Multi-Task | Exact Span F1 | 96.83 | — | Unverified |
| 6 | Flair | Exact Span F1 | 96.72 | — | Unverified |
| 7 | SeqVAT | Exact Span F1 | 95.45 | — | Unverified |
| 8 | Adversarial Training | Exact Span F1 | 95.25 | — | Unverified |
| 9 | BiLSTM-CRF | Exact Span F1 | 95.18 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ACE | F1 score | 97.3 | — | Unverified |
| 2 | Flair embeddings | F1 score | 96.72 | — | Unverified |
| 3 | JMT | F1 score | 95.77 | — | Unverified |
| 4 | Low supervision | F1 score | 95.57 | — | Unverified |
| 5 | IntNet + BiLSTM-CRF | F1 score | 95.29 | — | Unverified |
| 6 | Suzuki and Isozaki | F1 score | 95.15 | — | Unverified |
| 7 | NCRF++ | F1 score | 95.06 | — | Unverified |
| 8 | BI-LSTM-CRF (Senna) (ours) | F1 score | 94.46 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | ACE | F1 | 95 | — | Unverified |
| 2 | Wang et al., 2020 | F1 | 94.4 | — | Unverified |
| 3 | AIN | F1 | 94.04 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Wang et al., 2020 | F1 | 92 | — | Unverified |
| 2 | AIN | F1 | 91.71 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Def2Vec | AUC | 93.07 | — | Unverified |