Research map / Large language models / Language, safety and interpretability
Span Model for Open Information Extraction on Accurate Corpus
Junlang Zhan, Hai Zhao
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 |
|---|---|---|---|
| Use span selection instead of BIO sequence labeling | Open IE model · OIE2016 | AUC: 0.462 -> 0.489 | Improved |
| Use span selection instead of BIO sequence labeling | Open IE model · Re-OIE2016 | AUC: 0.579 -> 0.659 | Improved |
| Train on all OpenIE4 extractions weighted by confidence instead of filtering below 0.9 (ablation) | SpanOIE · OIE2016 | AUC: 0.491 -> 0.489 | No clear effect |
| Train on all OpenIE4 extractions weighted by confidence instead of filtering below 0.9 (ablation) | SpanOIE · Re-OIE2016 | AUC: 0.647 -> 0.659 | Improved |
| Train on all OpenIE4 extractions weighted by confidence instead of filtering below 0.9 (ablation) | SpanOIE · OIE2016 | F1: 69.42 -> 68.65 | Hurt |
| Train on all OpenIE4 extractions weighted by confidence instead of filtering below 0.9 (ablation) | SpanOIE · Re-OIE2016 | F1: 78.08 -> 78.50 | Improved |
| Add the parent-word BiLSTM feature for each span (ablation) | SpanOIE · OIE2016 | AUC: 0.489 -> 0.487 | Hurt |
| Add the parent-word BiLSTM feature for each span (ablation) | SpanOIE · Re-OIE2016 | AUC: 0.659 -> 0.639 | Hurt |
| Evaluate on manually relabeled Re-OIE2016 instead of OIE2016 | SpanOIE · Re-OIE2016 vs OIE2016 | AUC: 0.489 -> 0.659 | Improved |
| Evaluate on manually relabeled Re-OIE2016 instead of OIE2016 | SpanOIE · Re-OIE2016 vs OIE2016 | F1: 68.65 -> 78.50 | Improved |
Closest papers by meaning
- Small Language Model Makes an Effective Long Text Extractor (AAAI 2025 · similarity 0.86)
- Span Graph Transformer for Document-Level Named Entity Recognition (AAAI 2024 · similarity 0.85)
- An Autoregressive Text-to-Graph Framework for Joint Entity and Relation Extraction (AAAI 2024 · similarity 0.84)
- SeqGPT: An Out-of-the-Box Large Language Model for Open Domain Sequence Understanding (AAAI 2024 · similarity 0.84)
- SWEb: A Large Web Dataset for the Scandinavian Languages (ICLR 2025 · similarity 0.84)
- SoftMatcha: A Soft and Fast Pattern Matcher for Billion-Scale Corpus Searches (ICLR 2025 · similarity 0.84)
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.
