---
title: "Eduard Hovy — author profile and papers"
description: "Eduard Hovy, The University of Melbourne: 35 papers on Elamur Research (1992–2026), h-index 93 — affiliations, topics, co-authors and who cites them in…"
canonical: "https://research.elamur.ai/free/en/authors/eduard-hovy/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/eduard-hovy/)

[home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Eduard Hovy

# Eduard Hovy

Eduard Hovy works across a wide span of natural language processing, from knowledge-based machine translation in 1992 to neural models and resources for low-resource languages in the 2020s. Recurring threads in the papers here are coreference of entities and events, summarization and its evaluation, question answering, opinion analysis, semantic roles, and the construction of datasets and annotation schemes.

The most cited work in this library is [Hierarchical Attention Networks for Document Classification](https://md.research.elamur.ai/free/en/papers/yang-2016-hierarchical-attention-networks-document-classification/) (2016), which encodes words within sentences and sentences within documents, with attention at both levels. [Automatic Evaluation of Summaries Using N-gram Co-occurrence Statistics](https://md.research.elamur.ai/free/en/papers/lin-2003-automatic-evaluation-summaries-n-gram/) (2003), with [Chin-Yew Lin](https://md.research.elamur.ai/free/en/authors/chin-yew-lin/), asked whether summaries could be scored automatically as translation output was being scored with BLEU, and oriented the measure toward recall of reference content. [Determining the Sentiment of Opinions](https://md.research.elamur.ai/free/en/papers/kim-2004-determining-sentiment-opinions/) (2004) models an opinion as a topic, a holder, a claim and a sentiment. Other widely cited papers introduced the RACE reading comprehension dataset (2017) and unsupervised data augmentation for consistency training (2019).

The newest papers move in two directions. [CommonMorph](https://md.research.elamur.ai/free/en/papers/mahmudi-2026-commonmorph-participatory-morphological-documentation-platform/) (2026) is an open platform through which communities can document inflectional morphology for languages that lack such resources, and [Factuality Challenges in the Era of Large Language Models](https://md.research.elamur.ai/free/en/papers/augenstein-2023-factuality-challenges-era-large-language/) (2023) is a perspective on how chatbots affect the reliability of public information. A 2021 survey reviews data augmentation methods for NLP.

Hovy was long at the University of Southern California (1988–2018) and was an adjunct professor at the Language Technologies Institute of Carnegie Mellon University (2012–2024); Hovy has been at the University of Melbourne from 2022. Frequent co-authors include [Teruko Mitamura](https://md.research.elamur.ai/free/en/authors/teruko-mitamura/), [Zhengzhong Liu](https://md.research.elamur.ai/free/en/authors/zhengzhong-liu/) and [Marta Recasens](https://md.research.elamur.ai/free/en/authors/marta-recasens/); the most frequent topic is [coreference resolution](https://md.research.elamur.ai/free/en/topics/coreference-resolution/).

35 papers on this site · 1992–2026 · page 1 of 2

current affiliation [The University of Melbourne](https://md.research.elamur.ai/free/en/universities/the-university-of-melbourne/), Melbourne, Australia; [Parks Victoria](https://ror.org/025esa616), Melbourne, Australia

Eduard Hovy has 35 papers on this site, in 1992–2026, most of them on [Coreference resolution](https://md.research.elamur.ai/free/en/topics/coreference-resolution/), [Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/), [Semantic role labeling](https://md.research.elamur.ai/free/en/topics/semantic-role-labeling/); often with [Teruko Mitamura](https://md.research.elamur.ai/free/en/authors/teruko-mitamura/), [Zhengzhong Liu](https://md.research.elamur.ai/free/en/authors/zhengzhong-liu/), [Chin-Yew Lin](https://md.research.elamur.ai/free/en/authors/chin-yew-lin/). 8 of them are among the papers [Mova](https://md.research.elamur.ai/free/en/projects/mova/) reads.

[OpenAlex](https://openalex.org/A5060225743) · [ORCID](https://orcid.org/0000-0002-3270-7903) · [**h-index 93**](https://openalex.org/authors/A5060225743) · [690 works](https://openalex.org/works?filter=author.id:A5060225743) · [43 467 citations](https://openalex.org/works?filter=author.id:A5060225743&sort=cited_by_count:desc) OpenAlex, 2026-10-14

[35 works on this site](https://md.research.elamur.ai/free/en/authors/eduard-hovy/#papers) · [6 355 citations in our library](https://md.research.elamur.ai/free/en/authors/eduard-hovy/cited-by/) · 32 more works in our library

The profile is the library's match of the name; two people with one name are not always told apart.

Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Eduard+Hovy) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Eduard+Hovy) · [dblp](https://dblp.org/search?q=Eduard+Hovy) · [ACL Anthology](https://aclanthology.org/search/?q=Eduard+Hovy)

## Affiliations

- [The University of Melbourne](https://md.research.elamur.ai/free/en/universities/the-university-of-melbourne/) · Melbourne, Australia —current · 2022–2026 on 34 papers · employment (ORCID)
- [Parks Victoria](https://ror.org/025esa616) · Melbourne, Australia —current · 2026 on 1 paper
- [Carnegie Mellon University](https://md.research.elamur.ai/free/en/universities/carnegie-mellon-university/) · Pittsburgh, United States —2012–2025 on 221 papers · employment 2012–2024: Adjunct Professor, Language Technologies Institute (ORCID)
- [Robert Bosch (United States)](https://md.research.elamur.ai/free/en/companies/robert-bosch/) · Farmington Hills, United States —2023 on 2 papers
- [Allen Institute for Artificial Intelligence](https://md.research.elamur.ai/free/en/labs/allen-institute-for-artificial-intelligence/) · Seattle, United States —2019–2022 on 6 papers
- [Amazon (United States)](https://md.research.elamur.ai/free/en/companies/amazon/) · Seattle, United States —2020–2022 on 2 papers
- [Defense Advanced Research Projects Agency](https://ror.org/02caytj08) · Arlington, United States —2021–2022 on 2 papers
- [University of California, Berkeley](https://md.research.elamur.ai/free/en/universities/university-of-california-berkeley/) · Berkeley, United States —2020–2021 on 3 papers
- [University of Southern California](https://md.research.elamur.ai/free/en/universities/university-of-southern-california/) · Los Angeles, United States —1988–2018 on 185 papers
- [Universidad Nacional de Educación a Distancia](https://md.research.elamur.ai/free/en/universities/universidad-nacional-de-educacion-a-distancia/) · Madrid, Spain —2012–2015 on 2 papers · education (Wikidata)

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 19 more institutions on single bylines are not listed.

## Topics

[Coreference resolution 7](https://md.research.elamur.ai/free/en/topics/coreference-resolution/) · [Machine translation 3](https://md.research.elamur.ai/free/en/topics/machine-translation/) · [Semantic role labeling 2](https://md.research.elamur.ai/free/en/topics/semantic-role-labeling/) · [Question answering 2](https://md.research.elamur.ai/free/en/topics/question-answering/) · [Evaluation & benchmarks 2](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · [Reasoning 2](https://md.research.elamur.ai/free/en/topics/reasoning/) · [Knowledge graphs & ontologies 2](https://md.research.elamur.ai/free/en/topics/knowledge-graphs/) · [Discourse analysis 1](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · [Part-of-speech tagging 1](https://md.research.elamur.ai/free/en/topics/part-of-speech-tagging/) · [Language models & pretraining 1](https://md.research.elamur.ai/free/en/topics/language-models/) · [Information extraction 1](https://md.research.elamur.ai/free/en/topics/information-extraction/) · [Sentiment & emotion 1](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · [Summarization 1](https://md.research.elamur.ai/free/en/topics/summarization/) · [Morphology 1](https://md.research.elamur.ai/free/en/topics/morphology/) · [Semantic parsing & meaning representation 1](https://md.research.elamur.ai/free/en/topics/semantic-parsing/)

## Works by year

### 2020s

[**2026** 1 paper](https://md.research.elamur.ai/free/en/papers/mahmudi-2026-commonmorph-participatory-morphological-documentation-platform/) · [**2023** 1 paper](https://md.research.elamur.ai/free/en/papers/augenstein-2023-factuality-challenges-era-large-language/) · [**2021** 1 paper](https://md.research.elamur.ai/free/en/papers/feng-2021-survey-data-augmentation-approaches-nlp/)

### 2010s

[**2019** 2 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2019/) · [**2018** 1 paper](https://md.research.elamur.ai/free/en/papers/kang-2018-dataset-peer-reviews-peerread-collection/) · [**2017** 4 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2017/) · [**2016** 6 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2016/) · [**2014** 4 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2014/) · [**2013** 3 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2013/) · [**2010** 2 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2010/)

### 2000s

[**2008** 1 paper](https://md.research.elamur.ai/free/en/papers/kozareva-2008-semantic-class-learning-web-hyponym/) · [**2006** 1 paper](https://md.research.elamur.ai/free/en/papers/kim-2006-extracting-opinions-opinion-holders-topics/) · [**2004** 1 paper](https://md.research.elamur.ai/free/en/papers/kim-2004-determining-sentiment-opinions/) · [**2003** 2 papers](https://md.research.elamur.ai/free/en/authors/eduard-hovy/year/2003/) · [**2002** 1 paper](https://md.research.elamur.ai/free/en/papers/ravichandran-2002-learning-surface-text-patterns-question/) · [**2001** 1 paper](https://md.research.elamur.ai/free/en/papers/hovy-2001-semantics-based-answer-pinpointing/) · [**2000** 1 paper](https://md.research.elamur.ai/free/en/papers/lin-2000-automated-acquisition-topic-signatures-text/)

### 1990s

[**1994** 1 paper](https://md.research.elamur.ai/free/en/papers/hovy-1994-pangloss-knowledge-based-machine-translation/) · [**1992** 1 paper](https://md.research.elamur.ai/free/en/papers/hovy-1992-approximating-interlingua-principled-way/)

## Papers

The most cited first (OpenAlex and our library).

**1** [2](https://md.research.elamur.ai/free/en/authors/eduard-hovy/page/2/) [next →](https://md.research.elamur.ai/free/en/authors/eduard-hovy/page/2/)

1. **[Hierarchical Attention Networks for Document Classification](https://md.research.elamur.ai/free/en/papers/yang-2016-hierarchical-attention-networks-document-classification/)** *(2016 · 5 409 citations)* The paper looks at document classification, where a model must assign labels such as sentiment ratings or topics to long texts.
2. **[Automatic Evaluation of Summaries Using N-gram Co-occurrence Statistics](https://md.research.elamur.ai/free/en/papers/lin-2003-automatic-evaluation-summaries-n-gram/)** *(2003 · 1 967 citations)* Lin and Hovy ask whether summaries can be scored automatically, the way the machine translation community had begun using BLEU/NIST, instead of relying on expensive and inconsistent human judges.
3. **[Unsupervised Data Augmentation for Consistency Training](https://md.research.elamur.ai/free/en/papers/xie-2019-unsupervised-data-augmentation-consistency-training/)** *(2019 · 1 837 citations)* Deep networks usually need many labeled examples. Semi-supervised learning tries to make up for this with unlabeled data. One popular family of methods, consistency training, asks a model to predict the same output for an input and for a perturbed copy of it. These methods have mostly used weak perturbations, such as…
4. **[Determining the Sentiment of Opinions](https://md.research.elamur.ai/free/en/papers/kim-2004-determining-sentiment-opinions/)** *(2004 · 1 677 citations)* Kim and Hovy address opinion analysis. Starting from a collection of documents and a subject of interest, the goal is to find who holds an opinion about the topic and whether that opinion is positive, negative or neutral. They model an opinion as a tuple of Topic, Holder, Claim and Sentiment. The work proceeds in two…
5. **[RACE: Large-scale ReAding Comprehension Dataset From Examinations](https://md.research.elamur.ai/free/en/papers/lai-2017-race-large-scale-reading-comprehension/)** *(2017 · 1 619 citations)* RACE collects about 28,000 passages and 100,000 four-option questions from English exams for Chinese middle-school (RACE-M, ages 12–15) and high-school (RACE-H) students, written by teachers; options are human-written sentences, often not in the passage.
6. **[Learning surface text patterns for a Question Answering System](https://md.research.elamur.ai/free/en/papers/ravichandran-2002-learning-surface-text-patterns-question/)** *(2002 · 970 citations)* Open-domain QA systems usually depend on heavy external resources such as named entity taggers, WordNet, parsers and ontologies.
7. **[A Survey of Data Augmentation Approaches for NLP](https://md.research.elamur.ai/free/en/papers/feng-2021-survey-data-augmentation-approaches-nlp/)** *(2021 · 935 citations)* Data augmentation (DA) adds variety to training data without collecting new examples. It is standard in computer vision but less developed in NLP. Text is discrete, so continuous noise cannot be applied, and it is harder to keep labels unchanged after a transformation. Feng et al. present what they call the first…
8. **[End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF](https://md.research.elamur.ai/free/en/papers/ma-2016-end-end-sequence-labeling-bi/)** *(2016 · 852 citations)* Strong sequence labelers for tasks like POS tagging and NER have traditionally depended on manually designed features, gazetteers and task-specific preprocessing.
9. **[Visualizing and Understanding Neural Models in NLP](https://md.research.elamur.ai/free/en/papers/li-2016-visualizing-understanding-neural-models-nlp/)** *(2016 · 842 citations)* Neural NLP models often outperform feature-based systems, but their vector representations are hard to interpret.
10. **[Harnessing Deep Neural Networks with Logic Rules](https://md.research.elamur.ai/free/en/papers/hu-2016-harnessing-deep-neural-networks-logic/)** *(2016 · 619 citations)* Deep neural networks learn mostly from labeled examples. This makes them data-hungry, hard to interpret, and hard to steer with human knowledge. Hu et al. propose a general way to inject declarative first-order logic rules into arbitrary architectures such as CNNs and RNNs. Rules are written in soft logic, so truth…
11. **[The Automated Acquisition of Topic Signatures for Text Summarization](https://md.research.elamur.ai/free/en/papers/lin-2000-automated-acquisition-topic-signatures-text/)** *(2000 · 585 citations)* The paper addresses a core summarization problem: a text often describes the pieces of a complex concept (e.g., menu, eating, paying in a restaurant visit) without ever naming the concept itself.
12. **[Extracting Opinions, Opinion Holders, and Topics Expressed in Online News Media Text](https://md.research.elamur.ai/free/en/papers/kim-2006-extracting-opinions-opinion-holders-topics/)** *(2006 · 559 citations)* Kim and Hovy address a gap in opinion mining. Earlier systems could detect subjective words and their polarity, but they rarely identified who holds an opinion and what it is about. This matters especially for news, where many people, organizations and countries voice different views. The authors build a pipeline…
13. **[Learning Whom to Trust with MACE](https://md.research.elamur.ai/free/en/papers/hovy-2013-learning-whom-trust-mace/)** *(2013 · 472 citations)* Crowdsourcing platforms such as Amazon's Mechanical Turk (AMT) make it cheap to collect categorical labels for NLP.
14. **[The Creation and Analysis of a Website Privacy Policy Corpus](https://md.research.elamur.ai/free/en/papers/wilson-2016-creation-analysis-website-privacy-policy/)** *(2016 · 336 citations)* Website privacy policies are legally binding, yet most users neither read nor understand them, and regulators cannot review them at scale.
15. **[Scoring Coreference Partitions of Predicted Mentions: A Reference Implementation](https://md.research.elamur.ai/free/en/papers/pradhan-2014-scoring-coreference-partitions-predicted-mentions/)** *(2014 · 284 citations)* Coreference metrics such as MUC, B-cubed and CEAF were originally defined for gold mentions, and different research groups adapted them to predicted mentions in incompatible ways, often by adding or removing mentions artificially so that the key and response partitions line up.
16. **[A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications](https://md.research.elamur.ai/free/en/papers/kang-2018-dataset-peer-reviews-peerread-collection/)** *(2018 · 274 citations)* Quantitative research on scientific peer review has been difficult because reviews are usually visible only to journal editors and program chairs. This paper introduces PeerRead v1, the first openly released corpus of peer reviews intended for research. The authors drew on three sources:
17. **[Semantic Class Learning from the Web with Hyponym Pattern Linkage Graphs](https://md.research.elamur.ai/free/en/papers/kozareva-2008-semantic-class-learning-web-hyponym/)** *(2008 · 273 citations)* The paper tackles automatic construction of semantic lexicons, meaning lists of words that belong to a given category, using the web with almost no supervision.
18. **[Toward Semantics-Based Answer Pinpointing](https://md.research.elamur.ai/free/en/papers/hovy-2001-semantics-based-answer-pinpointing/)** *(2001 · 263 citations)* The paper tackles a weakness in factoid question answering (QA). Most systems pick the answer by sliding a word-scoring window over retrieved text. That approach cannot mark exact answer boundaries, cannot recognize the semantic type of an answer such as a person or a place, and fails when answer pieces are spread…
19. **[Shakespearizing Modern Language Using Copy-Enriched Sequence to Sequence Models](https://md.research.elamur.ai/free/en/papers/jhamtani-2017-shakespearizing-modern-language-copy-enriched/)** *(2017 · 251 citations)* The paper turns modern English into Shakespearean English with a sentence-level encoder-decoder that has a pointer (copy) component, so it can copy words from the input.
20. **[Factuality Challenges in the Era of Large Language Models](https://md.research.elamur.ai/free/en/papers/augenstein-2023-factuality-challenges-era-large-language/)** *(2023 · 241 citations)* This perspective article, written by researchers and fact-checking practitioners, looks at how chatbots built on LLMs such as ChatGPT, Bing Chat and Bard affect the reliability of public information. It presents no new dataset or model. Instead, it surveys existing evidence and frames the problem in two parts:
21. **[Squibs: What Is a Paraphrase?](https://md.research.elamur.ai/free/en/papers/bhagat-2013-squibs-what-is-paraphrase/)** *(2013 · 216 citations)* Paraphrase recognition matters for NLP tasks such as question answering, information extraction, textual entailment and machine translation.
22. **[Recursive Deep Models for Discourse Parsing](https://md.research.elamur.ai/free/en/papers/li-2014-recursive-deep-models-discourse-parsing/)** *(2014 · 206 citations)* The paper addresses document-level discourse parsing in the RST framework. In this task, a system must decide how clause-sized units join into a hierarchical tree and which rhetorical relation links each pair. The authors argue that hand-crafted lexical and syntactic features miss the semantic and intentional signals…
23. **[EQUATE: A Benchmark Evaluation Framework for Quantitative Reasoning in Natural Language Inference](https://md.research.elamur.ai/free/en/papers/ravichander-2019-equate-benchmark-evaluation-framework-quantitative/)** *(2019 · 122 citations)* EQUATE is a benchmark of five test sets for quantitative reasoning in entailment: RTE-Quant (166 pairs from RTE2–4), NewsNLI (968 crowdsourced pairs from news summaries), RedditNLI (250 expert-labelled headline pairs), a synthetic stress test (7,500 pairs) and AwpNLI (722 algebra word problems turned into pairs).
24. **[Supervised Within-Document Event Coreference using Information Propagation](https://md.research.elamur.ai/free/en/papers/liu-2014-supervised-within-document-event-coreference/)** *(2014 · 90 citations)* The paper addresses within-document event coreference, which means deciding when two event mentions in one text describe the same real-world event.
25. **[Maximum Entropy Models for FrameNet Classification](https://md.research.elamur.ai/free/en/papers/fleischman-2003-maximum-entropy-models-framenet-classification/)** *(2003 · 86 citations)* The paper tackles automatic semantic role labeling on FrameNet, where each annotated sentence has a target predicate whose frame elements must be assigned one of 120 semantic roles.

**1** [2](https://md.research.elamur.ai/free/en/authors/eduard-hovy/page/2/) [next →](https://md.research.elamur.ai/free/en/authors/eduard-hovy/page/2/)

## Also in our library

32 more works of theirs are in our library without a page here yet; they will appear as we publish them.

## Co-authors here

12 of 61 co-authors with a page, the most papers together first.

- [Teruko Mitamura](https://md.research.elamur.ai/free/en/authors/teruko-mitamura/) ×4
- [Zhengzhong Liu](https://md.research.elamur.ai/free/en/authors/zhengzhong-liu/) ×4
- [Chin-Yew Lin](https://md.research.elamur.ai/free/en/authors/chin-yew-lin/) ×3
- [Marta Recasens](https://md.research.elamur.ai/free/en/authors/marta-recasens/) ×3
- [Xuezhe Ma](https://md.research.elamur.ai/free/en/authors/xuezhe-ma/) ×3
- [Chris Dyer](https://md.research.elamur.ai/free/en/authors/chris-dyer/) ×2
- [Deepak Ravichandran](https://md.research.elamur.ai/free/en/authors/deepak-ravichandran/) ×2
- [Jiwei Li](https://md.research.elamur.ai/free/en/authors/jiwei-li/) ×2
- [Jun Araki](https://md.research.elamur.ai/free/en/authors/jun-araki/) ×2
- [Qizhe Xie](https://md.research.elamur.ai/free/en/authors/qizhe-xie/) ×2
- [Soo-Min Kim](https://md.research.elamur.ai/free/en/authors/soo-min-kim/) ×2
- [Varun Gangal](https://md.research.elamur.ai/free/en/authors/varun-gangal/) ×2

## Cited by

Papers of our library cite their works 6 355 times (of the 140 884 papers whose bibliographies we read); [553 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/eduard-hovy/cited-by/).