Research map / Architectures and efficiency / Attention and transformers
Jointly Parse and Fragment Ungrammatical Sentences
Homa B. Hashemi, Rebecca Hwa
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| seq2seq vs pipeline H&H16 for arc pruning | H&H16 · ESL arc pruning | UAS: 61.36 -> 82.4 | Improved |
| seq2seq vs retrained parser | RetrainedParser · ESL arc pruning | UAS: 63 -> 82.4 | Improved |
| seq2seq vs H&H16 on pruned arcs | H&H16 · ESL pruned-arc accuracy | F1: 0.48 -> 0.63 | Improved |
| seq2seq vs H&H16 on fragments | H&H16 · ESL tree fragments | set-2-set F1: 0.67 -> 0.83 | Improved |
| seq2seq vs RetrainedParser on fragments | RetrainedParser · ESL tree fragments | set-2-set F1: 0.77 -> 0.83 | Improved |
| fragment features from seq2seq vs H&H16 for binary grammaticality | H&H16 · FCE binary grammaticality | accuracy (%): 79.9 -> 81.3 | Improved |
| fragment features from seq2seq vs H&H16 for ordinal grammaticality | H&H16 · FCE ordinal grammaticality | Pearson r: 0.377 -> 0.377 | No clear effect |
| fragment features from RetrainedParser vs constituency baseline Post | Post · FCE binary grammaticality | accuracy (%): 77.3 -> 77.6 | No clear effect |
| apply seq2seq arc pruning vs no pruning | Basic · Semantic dependency detection | FDR: 12.81 -> 7.32 | Improved |
| apply H&H16 arc pruning vs no pruning | Basic · Semantic dependency detection | FDR: 12.81 -> 7.40 | Improved |
| apply RetrainedParser vs no pruning | Basic · Semantic dependency detection | FDR: 12.81 -> 7.88 | Improved |
| seq2seq vs RetrainedParser for argument errors | RetrainedParser · One-error sentences, argument error | FDR: 18.1 -> 16.7 | Improved |
Closest papers by meaning
- Global Greedy Dependency Parsing (AAAI 2020 · similarity 0.85)
- Dependency Parsing is More Parameter-Efficient with Normalization (NeurIPS 2025 · similarity 0.84)
- Improving Unsupervised Constituency Parsing via Maximizing Semantic Information (ICLR 2025 · similarity 0.84)
- Error Diversity Matters: An Error-Resistant Ensemble Method for Unsupervised Dependency Parsing (AAAI 2025 · similarity 0.83)
- Ensemble Distillation for Unsupervised Constituency Parsing (ICLR 2024 · similarity 0.83)
- Implicit Word Reordering with Knowledge Distillation for Cross-Lingual Dependency Parsing (AAAI 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
