ACL Findings2022
Revisiting the Effects of Leakage on Dependency Parsing
Nathaniel Krasner, Miriam Wanner, Antonios Anastasopoulos
Abstract
Recent work by Søgaard (2020) showed that, treebank size aside, overlap between training and test graphs (termed leakage) explains more of the observed variation in dependency parsing performance than other explanations. In this work we revisit this claim, testing it on more models and languages. We find that it only holds for zero-shot cross-lingual settings. We then propose a more fine-grained measure of such leakage which, unlike the original measure, not only explains but also correlates with observed performance variation. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on1
Related papers
- On the Relation between Syntactic Divergence and Zero-Shot PerformanceOfir Arviv, Dmitry Nikolaev, Taelin Karidi, Omri AbendEMNLP 2021 · 3 citations
- Genre as Weak Supervision for Cross-lingual Dependency ParsingMax Müller-Eberstein, Rob van der Goot, Barbara PlankEMNLP 2021 · 13 citations
- Annotations Matter: Leveraging Multi-task Learning to Parse UD and SUDZeeshan Ali Sayyed, Daniel DakotaACL Findings 2021 · 2 citations
- On the Branching Bias of Syntax Extracted from Pre-trained Language ModelsHuayang Li, Lemao Liu, Guoping Huang, Shuming ShiEMNLP Findings 2020 · 5 citations
- Implicit Word Reordering with Knowledge Distillation for Cross-Lingual Dependency ParsingZhuoran Li, Chunming Hu, Junfan Chen, Zhijun Chen et al.AAAI 2025 · 1 citation
