Massimo Poesio · page 3
Papers
- 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). - Combining Minimally-supervised Methods for Arabic Named Entity Recognition (2015 · 14 citations)
Supervised Arabic NER performs well, but every new domain or new entity class needs fresh expert annotation. - ARRAU: Linguistically-Motivated Annotation of Anaphoric Descriptions (2016 · 13 citations)
The paper presents the second release of ARRAU, a multi-domain corpus for anaphora. Its domains are news (from the RST Discourse Treebank), task-oriented dialogue (TRAINS), fiction (PEAR) and medical leaflets (GNOME). Release 2 has 552 documents, 348,072 tokens and 99,582 markables. Release 1 had 264 documents… - Aggregating Crowdsourced and Automatic Judgments to Scale Up a Corpus of Anaphoric Reference for Fiction and Wikipedia Texts (2023 · 12 citations)
Existing coreference corpora are mostly mid-sized, concentrated on news or science, short-document, and often skip singletons and non-referring expressions. - Neural Coreference Resolution for Arabic (2020 · 12 citations)
Arabic coreference resolution has barely advanced since the CoNLL 2012 shared task. The strongest system was still the feature-based resolver of Björkelund and Kuhn (2014), and no neural model had been tried. The authors fill this gap by adapting a leading English end-to-end architecture to Arabic. Their baseline is… - Neural Mention Detection (2020 · 11 citations)
Mention detection means finding every span that refers to an entity, including pronouns, common-noun phrases and nested spans, not only proper names. - The Universal Anaphora Scorer 2.0 (2023 · 10 citations)
The Universal Anaphora (UA) initiative wants evaluation tools that cover more than plain coreference. Version 1.0 of its scorer already handled identity anaphora, split antecedents, bridging and discourse deixis. However, it could only score contiguous mentions, and it could evaluate zero anaphora only when the zeros… - Anaphoric Zero Pronoun Identification: A Multilingual Approach (2020 · 10 citations)
Pro-drop languages such as Arabic and Chinese routinely omit subject pronouns that still refer back to earlier entities. - Scoring Coreference Chains with Split-Antecedent Anaphors (2023 · 9 citations)
Standard coreference scorers (MUC, B3, CEAF, LEA, BLANC) assume that every anaphor refers to one entity, so they cannot score split-antecedent references such as "they" for "John" and "Mary" in "John met Mary. - Referential ambiguity and clarification requests: comparing human and LLM behaviour (2025 · 4 citations)
This paper asks whether ambiguous references actually lead people to ask clarification questions, and whether LLMs behave the same way. - The ARRAU 3.0 Corpus (2024 · 4 citations)
ARRAU is a long-running effort, started in 2004, to build an English corpus whose anaphoric annotation goes well beyond plain identity coreference. - Universal Anaphora: The First Three Years (2024 · 4 citations)
Coreference work in NLP has long revolved around OntoNotes. That corpus centres on identity coreference in a limited set of genres and leaves out singletons, expletives and many harder phenomena. Universal Anaphora (UA) began in 2020 and is modelled on Universal Dependencies. It aims to coordinate research on… - Joint Coreference Resolution for Zeros and non-Zeros in Arabic (2022 · 4 citations)
In pro-drop languages such as Arabic, a subject can be left unspoken. These gaps, called anaphoric zero pronouns (AZPs), still refer back to earlier entities. Standard Arabic coreference systems cluster only overt mentions, and AZP resolvers link a zero to a single antecedent while ignoring the rest of the chain. The… - The QMUL/HRBDT contribution to the NADI Arabic Dialect Identification Shared Task (2020 · 4 citations)
The paper presents a system for the country-level subtask of the NADI 2020 shared task. The goal is to tell which of 21 Arab countries a tweet comes from, using noisy Twitter text with very uneven class sizes. The authors clean and normalise the tweets and up-sample minority classes. Each tweet is then represented by… - Coreference Annotation of an Arabic Corpus using a Virtual World Game (2022 · 3 citations)
Arabic coreference resources are small compared with those for English or Chinese. This limits how well resolvers can be trained. The authors present Stroll with a Scroll, a Game-With-A-Purpose (GWAP) that gathers Arabic coreference judgments from players. Unlike earlier coreference games, which were puzzle-style, it… - AmbiCoRefVis: A Tool for Visualizing Coreferential Ambiguity (2026)
Coreference annotation often involves genuine ambiguity: a pronoun may plausibly point to several antecedents, and annotators frequently disagree.
This page in Markdown → https://md.research.elamur.ai/free/en/authors/massimo-poesio/page/3/