Michael Strube · page 2
Papers
- Using Linguistic Features to Improve the Generalization Capability of Neural Coreference Resolvers (2018 · 30 citations)
Neural coreference resolvers that win on CoNLL-2012 lose their ranking on WikiCoref, which uses the same guidelines. - Bridging Textual Ellipses (1996 · 28 citations)
The paper addresses textual ellipsis (functional anaphora) in German. A definite noun phrase such as "Ladezeit" (charge time) refers back to an entity from the previous sentence (an accumulator) through an unstated conceptual relation, not through identity. A text understanding system must infer this link, or its… - Unrestricted Coreference Resolution via Global Hypergraph Partitioning (2011 · 27 citations)
The paper describes COPA, a coreference resolver entered in the open setting of the CoNLL-2011 shared task on unrestricted coreference over OntoNotes. - Using Automatically Extracted Minimum Spans to Disentangle Coreference Evaluation from Boundary Detection (2019 · 20 citations)
Scoring coreference by maximum spans mixes coreference errors with boundary errors such as PP attachment: «an extensive presence» and «an extensive presence, of course in this country» count as different mentions, and the system is penalised as if it had linked the wrong thing. - The CODI-CRAC 2022 Shared Task on Anaphora, Bridging, and Discourse Deixis in Dialogue (2022 · 14 citations)
The second edition of a shared task on anaphora in dialogue, with three tasks: identity anaphora (including split antecedents and singletons), bridging reference, and discourse deixis (a pronoun such as that or this referring to an event, fact or utterance). - Revisiting Selectional Preferences for Coreference Resolution (2017 · 14 citations)
Selectional preferences, meaning what kinds of arguments a predicate slot plausibly takes, have long been considered helpful for coreference. - Annotating Anaphoric and Bridging Relations with MMAX (2001 · 14 citations)
This paper addresses the need for manually annotated corpora to train and evaluate systems that resolve anaphoric and bridging relations. - Analyzing and Visualizing Coreference Resolution Errors (2015 · 13 citations)
Coreference resolution produces sets of mentions, so it is not obvious how to name the specific errors when a system's clusters differ from the gold ones. - What Causes the Failure of Explicit to Implicit Discourse Relation Recognition? (2024 · 12 citations)
Classifiers for implicit discourse relations are often trained on explicit examples with their connectives removed, but they do poorly on genuine implicit data. - Search Space Pruning: A Simple Solution for Better Coreference Resolvers (2016 · 11 citations)
Coreference resolvers perform far worse on automatically detected mentions than on gold mentions. With Cort's latent ranking model, the gap on CoNLL 2012 English is about 17 points. Moosavi and Strube attribute much of this to a large, unbalanced search space, since most noun-phrase mentions are singletons or… - Investigating Multilingual Coreference Resolution by Universal Annotations (2023 · 6 citations)
The paper asks whether the harmonized annotations in CorefUD can reveal properties of coreference shared across languages, and whether such properties help a multilingual coreference resolution (MCR) system.
This page in Markdown → https://md.research.elamur.ai/free/en/authors/michael-strube/page/2/