---
title: "An Annotated Corpus of Reference Resolution for Interpreting Common Grounding (2019)"
description: "Common grounding, meaning how dialogue partners build, fix and revise shared understanding, is hard to inspect. 36 references, cited by 6 papers here."
authors: ["Takuma Udagawa", "Akiko Aizawa"]
year: 2019
venue: "arXiv:1911.07588"
source: "https://arxiv.org/abs/1911.07588"
doi: "10.48550/arxiv.1911.07588"
arxiv: "1911.07588"
licence: "arXiv non-exclusive distribution licence (read only, no redistribution)"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/udagawa-2019-annotated-corpus-reference-resolution-interpreting/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/udagawa-2019-annotated-corpus-reference-resolution-interpreting/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2019](https://md.research.elamur.ai/free/en/papers/2019/page/6/) / An Annotated Corpus of Reference Resolution for Interpreting Common Grounding

# An Annotated Corpus of Reference Resolution for Interpreting Common Grounding

**Authors:** [Takuma Udagawa](https://md.research.elamur.ai/free/en/authors/takuma-udagawa/) and [Akiko Aizawa](https://md.research.elamur.ai/free/en/authors/akiko-aizawa/) · **Year:** 2019 · **Venue:** [arXiv:1911.07588](https://md.research.elamur.ai/free/en/venues/proceedings-of-the-aaai-conference-on-artificial-intelligence/) ([Proceedings of the AAAI Conference on Artificial Intelligence 2019](https://md.research.elamur.ai/free/en/venues/proceedings-of-the-aaai-conference-on-artificial-intelligence/2019/)) · **Publisher:** [Association for the Advancement of Artificial Intelligence](https://md.research.elamur.ai/free/en/publishers/association-for-the-advancement-of-artificial-intelligence/) · **Licence:** [arXiv non-exclusive distribution licence (read only, no redistribution)](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/1911.07588) · [PDF](https://arxiv.org/pdf/1911.07588) · [DOI 10.48550/arxiv.1911.07588](https://doi.org/10.48550/arxiv.1911.07588) · [arXiv 1911.07588](https://arxiv.org/abs/1911.07588) · [OpenAlex](https://openalex.org/W2998108220)

**Citations:** 8 citations (OpenAlex, 2026-10-14) · [6](https://md.research.elamur.ai/free/en/papers/udagawa-2019-annotated-corpus-reference-resolution-interpreting/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
Common grounding, meaning how dialogue partners build, fix and revise shared understanding, is hard to inspect. This is especially true when the shared context is continuous and only partly visible to each side, and when the speaker is an end-to-end neural system. The authors treat reference resolution as the core subtask of grounding and annotate it directly. They take the 5,191 successful dialogues of the OneCommon task, in which two players see overlapping views of dots and must agree on a shared one. They detect referring expressions (markables) and ask several non-expert annotators to mark which dots each expression points to, from the speaker's view. Agreement is high, and the remaining disagreements tend to reflect genuine ambiguity. Baseline neural models trained jointly on reference resolution and the end task do better at target selection and selfplay. The annotation also shows where the models fail: they often choose the correct dot without having tracked the referents correctly. The resource makes grounding measurable step by step, for both humans and machines.

## Key points
- **Corpus size:** 40,172 referring expressions across 5,191 dialogues from the OneCommon corpus (6,760 dialogues in total). Each referent interpretation has at least 3 judgements. The vocabulary after cleaning is small, with only 2,035 unique tokens.
- **Task setup:** each player sees 7 dots, of which 4, 5 or 6 are shared. Dots differ only in continuous properties: color, size and position.
- **Markable definition:** a minimal noun phrase that keeps prenominal modifiers (determiners, quantifiers, adjectives) and drops postnominal ones (PPs, relative clauses), so spans never nest or overlap. Optional flags mark generic, all-referents and no-referent cases.
- **Baseline models:** five variants (TSEL, REF, TSEL-REF, TSEL-DIAL, TSEL-REF-DIAL), each run 10 times on 8:1:1 splits. Adding reference resolution raises target selection from 67.79 to about 69. It also raises selfplay success in every setting, for example from 42.07 to 45.78 with 4 shared dots.
- **Reference resolution results:**
  - The models reach about 85.8% entity-level accuracy but only about 33–34% exact match. Humans reach 96.26 and 86.90 respectively.
  - Exact match falls sharply when an expression has more than one referent: 36.86% for one referent versus 20.59% for two.
  - Reference resolution and target selection scores correlate only weakly, at 0.23.
  - A BiLSTM-CRF markable tagger trained on the corpus reaches 98.9% token accuracy.

## Interesting ideas
- Restricting mention spans to the head noun plus its left-side modifiers guarantees that spans never nest, which makes annotation and evaluation much simpler.
- Reference is annotated as a direct link from an expression to a set of world entities, not as a chain of mentions, so plurals, groupings and part-of relations come along for free.
- Scoring the subtask alongside the end task, and checking how strongly they correlate, exposes answers that are right for the wrong reason.
- Collecting several judgements and keeping their reasonable disagreement treats ambiguity as signal rather than noise.
- Breaking results down by how many referents an expression has shows that multi-referent tracking is the real weak point, which a single averaged score hides.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (frames): coreference mentions could be cut from the UD tree as the noun plus its det/amod/nummod dependents, leaving nmod/acl outside the span to avoid nested-mention conflicts.
- **I2** (world): mentions could resolve straight to entity sets in the text's world model, with plural or grouped phrases mapping to explicit subsets that can be inspected.
- **I3** (answer): for each QA or fable question, log whether the entities behind the answer were resolved correctly, and flag correct answers built on wrong resolution as hidden failures.
- **I4** (learn): when witness parsers disagree on silver data, record that disagreement as a confidence or ambiguity label instead of forcing a single gold choice.
- **I5** (learn): evaluation reports could be split by the number of entities a phrase denotes, so that weak plural or group handling is caught early.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [The University of Tokyo](https://md.research.elamur.ai/free/en/universities/the-university-of-tokyo/) — University, Tokyo, Japan · 45 works · 8,692 citations
- [National Institute of Informatics](https://md.research.elamur.ai/free/en/labs/national-institute-of-informatics/) — Research institute, Tokyo, Japan · 21 works · 3,045 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works

## Cite

Cite as: Takuma Udagawa and Akiko Aizawa. 2019. An Annotated Corpus of Reference Resolution for Interpreting Common Grounding. arXiv:1911.07588.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{udagawa2019annotated,
  title = {{An Annotated Corpus of Reference Resolution for Interpreting Common Grounding}},
  author = {Takuma Udagawa and Akiko Aizawa},
  year = {2019},
  doi = {10.48550/arxiv.1911.07588},
  eprint = {1911.07588},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.1911.07588}
}
```

- [← previous Squared English Word: A Method of Generating Glyph to Use Super Characters for Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/sun-2019-squared-english-word-method-generating/)
- [next → Modeling Long-Range Context for Concurrent Dialogue Acts Recognition](https://md.research.elamur.ai/free/en/papers/yu-2019-modeling-long-range-context-concurrent/)
- [↑ back to papers of 2019](https://md.research.elamur.ai/free/en/papers/2019/page/6/)

## References

36 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Bahdanau, D.; Cho, K.; and Bengio, Y. 2014. Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473. [↗](https://arxiv.org/abs/1409.0473) [our card](https://md.research.elamur.ai/free/en/papers/bahdanau-2014-neural-machine-translation-jointly-learning/)
2. Belinkov, Y., and Glass, J. 2019. Analysis methods in neural language processing: A survey. Transactions of the Association for Computational Linguistics 7:49–72. [↗](https://doi.org/10.1162/tacl_a_00254) [our card](https://md.research.elamur.ai/free/en/papers/belinkov-2019-analysis-methods-neural-language-processing/)
3. Bordes, A., and Weston, J. 2016. Learning end-to-end goal-oriented dialog. CoRR abs/1605.07683. [↗](https://arxiv.org/abs/1605.07683)
4. Chen, H.; Suhr, A.; Misra, D.; Snavely, N.; and Artzi, Y. 2019. Touchdown: Natural language navigation and spatial reasoning in visual street environments. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 12538–12547. [find](https://scholar.google.com/scholar?q=Chen%2C+H%2E%3B+Suhr%2C+A%2E%3B+Misra%2C+D%2E%3B+Snavely%2C+N%2E%3B+and+Artzi%2C+Y%2E+2019%2E+Touchdown%3A+Natural+language+navigation+and+spatial+reasoning+in+visual+street+environments%2E+In+Proceedings+of+the+IEEE+Conference+on+Com)
5. Cho, K.; van Merrienboer, B.; Bahdanau, D.; and Bengio, Y. 2014. On the properties of neural machine translation: Encoder–decoder approaches. In Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, 103–111. Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/cho-2014-properties-neural-machine-translation-encoder/)
6. Clark, H. H. 1996. Using language. Cambridge university press. [find](https://scholar.google.com/scholar?q=Clark%2C+H%2E+H%2E+1996%2E+Using+language%2E+Cambridge+university+press%2E)
7. Das, A.; Kottur, S.; Gupta, K.; Singh, A.; Yadav, D.; Moura, J. M.; Parikh, D.; and Batra, D. 2017. Visual dialog. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, volume 2. [find](https://scholar.google.com/scholar?q=Das%2C+A%2E%3B+Kottur%2C+S%2E%3B+Gupta%2C+K%2E%3B+Singh%2C+A%2E%3B+Yadav%2C+D%2E%3B+Moura%2C+J%2E+M%2E%3B+Parikh%2C+D%2E%3B+and+Batra%2C+D%2E+2017%2E+Visual+dialog%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition%2C+vol)
8. De Vries, H.; Strub, F.; Chandar, S.; Pietquin, O.; Larochelle, H.; and Courville, A. 2017. Guesswhat?! visual object discovery through multi-modal dialogue. In Proc. of CVPR. [↗](https://doi.org/10.48550/arxiv.1611.08481)
9. Doshi-Velez, F., and Kim, B. 2017. Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608. [↗](https://arxiv.org/abs/1702.08608) [our card](https://md.research.elamur.ai/free/en/papers/doshi-velez-2017-towards-rigorous-science-interpretable-machine/)
10. Fleiss, J. L. 1971. Measuring nominal scale agreement among many raters. Psychological bulletin 76(5):378. [↗](https://doi.org/10.1037/h0031619)
11. Götze, J., and Boye, J. 2016. SpaceRef: A corpus of street-level geographic descriptions. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), 3822–3827. Portorož, Slovenia: European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Go%CC%88tze%2C+J%2E%2C+and+Boye%2C+J%2E+2016%2E+SpaceRef%3A+A+corpus+of+street%2Dlevel+geographic+descriptions%2E+In+Proceedings+of+the+Tenth+International+Conference+on+Language+Resources+and+Evaluation+%28LREC%E2%80%9916%29%2C+3822%E2%80%93382)
12. Gururangan, S.; Swayamdipta, S.; Levy, O.; Schwartz, R.; Bowman, S.; and Smith, N. A. 2018. Annotation artifacts in natural language inference data. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers), 107–112. New Orleans, Louisiana: Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Gururangan%2C+S%2E%3B+Swayamdipta%2C+S%2E%3B+Levy%2C+O%2E%3B+Schwartz%2C+R%2E%3B+Bowman%2C+S%2E%3B+and+Smith%2C+N%2E+A%2E+2018%2E+Annotation+artifacts+in+natural+language+inference+data%2E+In+Proceedings+of+the+2018+Conference+of+the+North+)
13. Haber, J.; Baumgärtner, T.; Takmaz, E.; Gelderloos, L.; Bruni, E.; and Fernández, R. 2019. The PhotoBook dataset: Building common ground through visually-grounded dialogue. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 1895–1910. Florence, Italy: Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/haber-2019-photobook-dataset-building-common-ground/)
14. Hernndez-Orallo, J. 2017. The Measure of All Minds: Evaluating Natural and Artificial Intelligence. New York, NY, USA: Cambridge University Press, 1st edition. [find](https://scholar.google.com/scholar?q=Hernndez%2DOrallo%2C+J%2E+2017%2E+The+Measure+of+All+Minds%3A+Evaluating+Natural+and+Artificial+Intelligence%2E+New+York%2C+NY%2C+USA%3A+Cambridge+University+Press%2C+1st+edition%2E)
15. Huang, Z.; Xu, W.; and Yu, K. 2015. Bidirectional lstm-crf models for sequence tagging. CoRR abs/1508.01991. [↗](https://arxiv.org/abs/1508.01991) [our card](https://md.research.elamur.ai/free/en/papers/huang-2015-bidirectional-lstm-crf-models-sequence/)
16. Ilinykh, N.; Zarrieß, S.; and Schlangen, D. 2019. Meetup! a corpus of joint activity dialogues in a visual environment. arXiv preprint arXiv:1907.05084. [↗](https://arxiv.org/abs/1907.05084)
17. Koschmann, T., and LeBaron, C. D. 2003. Reconsidering common ground. In ECSCW 2003, 81–98. Springer. [↗](https://doi.org/10.1007/978-94-010-0068-0_5)
18. Linzen, T.; Dupoux, E.; and Goldberg, Y. 2016. Assessing the ability of lstms to learn syntax-sensitive dependencies. Transactions of the Association for Computational Linguistics 4:521–535. [↗](https://doi.org/10.1162/tacl_a_00115) [our card](https://md.research.elamur.ai/free/en/papers/linzen-2016-assessing-ability-lstms-learn-syntax/)
19. Lipton, Z. C. 2016. The mythos of model interpretability. arXiv preprint arXiv:1606.03490. [↗](https://arxiv.org/abs/1606.03490)
20. McCoy, T.; Pavlick, E.; and Linzen, T. 2019. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 3428–3448. Florence, Italy: Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/mccoy-2019-right-wrong-reasons-diagnosing-syntactic/)
21. Poesio, M., and Rieser, H. 2010. Completions, coordination, and alignment in dialogue. Dialogue & Discourse 1(1). [↗](https://doi.org/10.5087/dad.2010.001) [our card](https://md.research.elamur.ai/free/en/papers/poesio-2010-completions-coordination-alignment-dialogue/)
22. Poesio, M.; Chamberlain, J.; Paun, S.; Yu, J.; Uma, A.; and Kruschwitz, U. 2019. A crowdsourced corpus of multiple judgments and disagreement on anaphoric interpretation. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 1778–1789. Minneapolis, Minnesota: Association for Computational Linguistics. [↗](https://doi.org/10.5283/epub.43420) [our card](https://md.research.elamur.ai/free/en/papers/poesio-2019-crowdsourced-corpus-multiple-judgments-disagreement/)
23. Poesio, M.; Stuckardt, R.; and Versley, Y. 2016. Anaphora resolution. Springer. [find](https://scholar.google.com/scholar?q=Poesio%2C+M%2E%3B+Stuckardt%2C+R%2E%3B+and+Versley%2C+Y%2E+2016%2E+Anaphora+resolution%2E+Springer%2E)
24. Potts, C. 2012. Goal-driven answers in the cards dialogue corpus. In Proceedings of the 30th west coast conference on formal linguistics, 1–20. Cascadilla Proceedings Project. [find](https://scholar.google.com/scholar?q=Potts%2C+C%2E+2012%2E+Goal%2Ddriven+answers+in+the+cards+dialogue+corpus%2E+In+Proceedings+of+the+30th+west+coast+conference+on+formal+linguistics%2C+1%E2%80%9320%2E+Cascadilla+Proceedings+Project%2E)
25. Pradhan, S.; Ramshaw, L.; Marcus, M.; Palmer, M.; Weischedel, R.; and Xue, N. 2011. Conll-2011 shared task: Modeling unrestricted coreference in ontonotes. In Proceedings of the Fifteenth Conference on Computational Natural Language Learning: Shared Task, 1–27. Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/pradhan-2011-conll-2011-shared-task-modeling/)
26. Recasens, M.; Martı́, M. A.; and Orasan, C. 2012. Annotating near-identity from coreference disagreements. In Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC-2012), 165–172. Istanbul, Turkey: European Languages Resources Association (ELRA). [our card](https://md.research.elamur.ai/free/en/papers/recasens-2012-annotating-near-identity-coreference-disagreements/)
27. Sankar, C.; Subramanian, S.; Pal, C.; Chandar, S.; and Bengio, Y. 2019. Do neural dialog systems use the conversation history effectively? an empirical study. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, 32–37. Florence, Italy: Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Sankar%2C+C%2E%3B+Subramanian%2C+S%2E%3B+Pal%2C+C%2E%3B+Chandar%2C+S%2E%3B+and+Bengio%2C+Y%2E+2019%2E+Do+neural+dialog+systems+use+the+conversation+history+effectively%3F+an+empirical+study%2E+In+Proceedings+of+the+57th+Annual+Meeting)
28. Shore, T.; Androulakaki, T.; and Skantze, G. 2018. KTH tangrams: A dataset for research on alignment and conceptual pacts in task-oriented dialogue. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). Miyazaki, Japan: European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Shore%2C+T%2E%3B+Androulakaki%2C+T%2E%3B+and+Skantze%2C+G%2E+2018%2E+KTH+tangrams%3A+A+dataset+for+research+on+alignment+and+conceptual+pacts+in+task%2Doriented+dialogue%2E+In+Proceedings+of+the+Eleventh+International+Confer)
29. Stenetorp, P.; Pyysalo, S.; Topić, G.; Ohta, T.; Ananiadou, S.; and Tsujii, J. 2012. Brat: a web-based tool for nlp-assisted text annotation. In Proceedings of the Demonstrations at the 13th Conference of the European Chapter of the Association for Computational Linguistics, 102–107. Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/stenetorp-2012-brat-web-based-tool-nlp/)
30. Sugawara, S.; Inui, K.; Sekine, S.; and Aizawa, A. 2018. What makes reading comprehension questions easier? In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 4208–4219. Brussels, Belgium: Association for Computational Linguistics. [our card](https://md.research.elamur.ai/free/en/papers/sugawara-2018-what-makes-reading-comprehension-questions/)
31. Sugawara, S.; Yokono, H.; and Aizawa, A. 2017. Prerequisite skills for reading comprehension: Multi-perspective analysis of mctest datasets and systems. In AAAI, 3089–3096. [find](https://scholar.google.com/scholar?q=Sugawara%2C+S%2E%3B+Yokono%2C+H%2E%3B+and+Aizawa%2C+A%2E+2017%2E+Prerequisite+skills+for+reading+comprehension%3A+Multi%2Dperspective+analysis+of+mctest+datasets+and+systems%2E+In+AAAI%2C+3089%E2%80%933096%2E)
32. Tokunaga, T.; Iida, R.; Terai, A.; and Kuriyama, N. 2012. The REX corpora: A collection of multimodal corpora of referring expressions in collaborative problem solving dialogues. In Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC’12), 422–429. Istanbul, Turkey: European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Tokunaga%2C+T%2E%3B+Iida%2C+R%2E%3B+Terai%2C+A%2E%3B+and+Kuriyama%2C+N%2E+2012%2E+The+REX+corpora%3A+A+collection+of+multimodal+corpora+of+referring+expressions+in+collaborative+problem+solving+dialogues%2E+In+Proceedings+of+the)
33. Traum, D. R. 1994. A computational theory of grounding in natural language conversation. Technical report, ROCHESTER UNIV NY DEPT OF COMPUTER SCIENCE. [find](https://scholar.google.com/scholar?q=Traum%2C+D%2E+R%2E+1994%2E+A+computational+theory+of+grounding+in+natural+language+conversation%2E+Technical+report%2C+ROCHESTER+UNIV+NY+DEPT+OF+COMPUTER+SCIENCE%2E)
34. Udagawa, T., and Aizawa, A. 2019. A natural language corpus of common grounding under continuous and partially-observable context. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, 7120–7127. [our card](https://md.research.elamur.ai/free/en/papers/udagawa-2019-natural-language-corpus-common-grounding/)
35. Vinyals, O., and Le, Q. V. 2015. A neural conversational model. CoRR abs/1506.05869. [↗](https://arxiv.org/abs/1506.05869) [our card](https://md.research.elamur.ai/free/en/papers/vinyals-2015-neural-conversational-model/)
36. Zarrieß, S.; Hough, J.; Kennington, C.; Manuvinakurike, R.; De-Vault, D.; Fernández, R.; and Schlangen, D. 2016. PentoRef: A corpus of spoken references in task-oriented dialogues. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), 125–131. Portorož, Slovenia: European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Zarrie%C3%9F%2C+S%2E%3B+Hough%2C+J%2E%3B+Kennington%2C+C%2E%3B+Manuvinakurike%2C+R%2E%3B+De%2DVault%2C+D%2E%3B+Ferna%CC%81ndez%2C+R%2E%3B+and+Schlangen%2C+D%2E+2016%2E+PentoRef%3A+A+corpus+of+spoken+references+in+task%2Doriented+dialogues%2E+In+Proceedings+of+)

*Updated 24 October 2026*