Machine Understanding of Scientific Language

Dustin Wright

REVIEW 2 major objections 5 minor 286 references

Machine Understanding of Scientific Language

T0 review · 2 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This thesis argues that faithfulness of science communication should be measured as degrees of information change between paired statements, and shows that models trained on a new dataset can reveal large-scale trends in that change.

desk verdict SPICED is a genuine new resource for measuring information change in science communication, but the large-scale trend analyses in Section 9.6 need a pre-filtering bias check before they can be trusted. read the letter →

arxiv 2506.23990 v1 pith:WD45SZXX submitted 2025-06-30 cs.CL cs.LG

Dustin Wright This is my paper

classification cs.CLcs.LG

keywords scientificmisinformationsciencecommunicationinformationchangesemanticmatchingfactcheckingSPICEDfew-shotlearningnaturallanguageprocessing

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The thesis tries to establish that the fidelity of scientific communication can be treated as a measurable, continuous quantity rather than a binary true/false label. The author argues that most scientific misinformation is not flat contradiction but subtle change—exaggeration, softened certainty, dropped caveats—and that the right unit of analysis is a pair of statements that express the same scientific finding. To make this concrete, the thesis builds SPICED, a dataset that pairs findings from papers, news articles, and Twitter and scores each pair on a 1–5 Information Matching Scale, then trains models that assign such scores to unseen pairs. If the claim is right, these models become an instrument for auditing science communication at scale: comparing outlets, accounts, and paper sections, and supplying evidence for scientific fact checking. The result matters because it gives researchers and eventually the public a way to say not just 'false' but 'this is how far the message moved from the source.'

What carries the argument

The machinery is the pair itself: a matched (source paper statement, media statement) scored by the Information Matching Score (IMS), a 1–5 scale that records whether the two texts express the same scientific finding and how much the information changed. SPICED operationalizes this by using sentence-embedding similarity to propose candidate pairs, then having domain-expert annotators score them; models fine-tuned on this supervision become the measurement instrument. This replaces the categorical verdict of fact checking (supports/refutes) with a graded score that can be thresholded for matching or averaged over groups to reveal trends.

What would settle it

Give the system statements drawn from papers that were later retracted or contradicted; if the models rate faithful restatements of those papers as low information change while independent human judges call them false, the ground-truth assumption fails in a measurable way.

Watch

Extended reading notes

Core claim

The central discovery, on the paper's own terms, is that information change in science communication can be operationalized as a supervised measurement task. The thesis defines a scientific finding as a statement describing a research output, then defines the Information Matching Score (IMS) as a 5-point rating of whether two statements describe the same finding and how much the information has changed. Using sentence embeddings to propose candidate pairs and domain experts to label them, the thesis constructs SPICED across three stages of communication—papers, news, and Twitter—and four fields, and shows that models fine-tuned on it outperform general semantic similarity models at scoring information change. It then demonstrates two uses of the resulting instrument: zero-shot evidence retrieval for scientific fact checking, and large-scale analyses showing that general news outlets change scientific messages more than science-specific outlets, and that limitations sections are among the most exaggerated parts of a paper.

Load-bearing premise

The whole framework treats scientific papers as ground truth, so a media statement that faithfully repeats a wrong or soon-retracted paper gets a low information-change score and escapes the misinformation flag.

Editorial extensions

If this is right

Faithfulness becomes a number: any new statement about a finding can be scored for how much it departs from the paper, enabling ranking and thresholding rather than binary accusation.
The same pairwise matching model transfers to evidence retrieval for scientific fact checking, improving performance without additional task-specific training data.
Aggregate measurement becomes possible: media outlets, social media accounts, and paper sections can be compared by average information change, exposing where distortion concentrates.
Because the dataset spans papers, news, and Twitter across four scientific fields, tools built on it are not locked to a single domain or genre.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural extension the thesis sketches but does not build is a taxonomy of change types—exaggeration, certainty shift, cherry-picking, flat contradiction—so that a score can be accompanied by an explanation of what changed.
The instrument could be repackaged as a proofreading aid for science journalists, flagging draft sentences whose implied information change exceeds a chosen threshold relative to the cited paper.
Since the ground truth is the paper itself, combining the score with external signals such as retraction lists or replication outcomes would separate 'faithful to a correct paper' from 'faithful to a wrong paper,' a distinction the thesis explicitly leaves open.
The same matching-plus-scoring recipe is testable in other languages and on preprint-to-published-version pairs, which would show whether information change is driven by editorial process, translation, or field norms.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read this for Chapter 9. The SPICED dataset is the real contribution: six thousand paper/news/tweet pairs scored on a five-point Information Matching Score, spanning four fields. The reframing of science communication quality as degree of information change, decoupled from binary veracity, is genuinely useful, and the thesis is honest about its scope. The individual chapters are peer-reviewed, the code and data links are there, and the experiments are multi-seed. That is more than most compilations offer.

The soft spot is the one the stress-test flags. SPICED candidates were pre-filtered by SentenceBERT, so annotators only saw pairs that SBERT thought were similar. The model trained on that distribution is then used in Section 9.6 both to retrieve matched findings and to rank media outlets, verified accounts, and paper sections. If SBERT similarity correlates with outlet style or register, the ranked differences could be selection artifacts. The thesis does not report coverage over SBERT similarity bands or validate the rankings on pairs that would have failed the pre-filter. That is the least secure link in the central claim. It is fixable: stratified re-annotation or a retrieval-recall check on a sample of unfiltered pairs would tell you whether the comparisons survive.

The smaller caveats are the small annotation sizes and the acknowledged papers-as-truth assumption. The latter is real but mainly limits normative conclusions about misinformation; the pre-filtering issue threatens the measurement itself. Chapter 7's generated claims come from the same corpus as the test claims, which is a data-construction caveat, not circularity.

This is not a takedown. For someone working on scientific fact checking or science communication, SPICED and the information-change framing are worth engaging with seriously, and the paper deserves referee time. But I would not take the Section 9.6 rankings at face value until the selection bias check is done.

Referee Report

2 major / 5 minor

Summary. The thesis compiles a PhD dissertation on machine understanding of scientific language, organized around three areas: automated fact checking, learning with limited data, and scientific text processing. It develops methods and datasets for check-worthiness detection, adversarial claim generation, multi-source domain adaptation, learning from crowd labels, cite-worthiness detection, zero-shot scientific fact checking, exaggeration detection, and, as the culminating contribution, a new task of modeling information change in science communication. The central thesis-level claim is that information change can be operationalized through the SPICED dataset and Information Matching Score (IMS), and that models trained on SPICED can be used to reveal large-scale trends in how scientific messages change across papers, news, and social media.

Significance. If the central claim holds, the thesis provides a measurement instrument and benchmark that decouples the study of scientific misinformation from binary veracity, offering a new way to quantify subtle distortions such as exaggeration and hedging across the science communication pipeline. The individual chapters have been peer-reviewed, use multi-seed experiments, and provide code/data links, which supports the reliability of the component empirical claims. The thesis is also commendable for stating key limitations explicitly, particularly the assumption that scientific documents represent truth. However, the significance of the synthesis-level claim depends on whether the SPICED-based models generalize to pairs that the SentenceBERT pre-filter would have excluded; this is the least secure link and needs additional validation before the large-scale trend analyses can be taken as measuring information change rather than retrieval artifacts.

major comments (2)

[§9.3.1–§9.3.3; §1.4.3] The load-bearing condition for the thesis-level claim is that models trained on SPICED measure information change on the same distribution of sentence pairs encountered in the wild. SPICED is constructed by first using SentenceBERT to select candidate pairs and then asking annotators to label only those candidates (§9.3.1–§9.3.3; summarized in §1.4.3). This conditions the IMS training distribution on SBERT similarity: same-finding paraphrases that are lexically or semantically distant are systematically under-sampled, and the proportion of hard candidates may differ by media source. The large-scale applications (§9.6) then use a model trained on this filtered distribution to retrieve matched findings and score information change across outlets and accounts. If SBERT similarity correlates with outlet style or register, the reported rankings (general news > press releases/science news; verified users > organizational accounts) could reflect retrieval or selection artifacts rather than actual information change. The thesis does not report stratified coverage over SBERT similarity bands or a validation of outlet-level rankings on pairs that would have failed the pre-filter. I therefore request a stratified analysis of SPICED coverage over SBERT similarity bands per source, and a re-annotation study of a random sample of pairs below the pre-filter threshold, with a comparison of the resulting IMS distributions and outlet rankings.
[§1.5.3] The thesis explicitly acknowledges in §1.5.3 that it assumes scientific documents represent truth. This is appropriately candid, but the framing in the abstract and §1.1.1—that the system can help identify misinformative scientific statements—goes beyond fidelity to papers. A statement can receive a low information-change score while the underlying paper is itself inaccurate, so the normative conclusions about misinformation require either a separate validation of paper accuracy or a systematic rewording of the claims to 'changes relative to the source paper.' The measurement framework itself can survive this limitation, but the societal-impact framing should be aligned with what is actually measured.

minor comments (5)

[Acknowledgements] The sentence 'who's insight, advice, and fine-tuned ability' should read 'whose insight, advice, and fine-tuned ability.'
[§4.4.1] The Amazon product review dataset description says '8,000 total tweets'; these are product reviews, not tweets, and the text should be corrected to 'reviews.'
[§1.4.3] The acronym SPICED is rendered as 'S PICED' in the running text, and the expansion 'SCIENTIFIC PARAPHRASE AND INFORMATION CHANG E D ATASET' contains a spurious line break; both should be corrected to 'SPICED' and 'CHANGE DATASET.'
[§2.4.5] The manual relabeling analysis uses only the top 100 predictions and does not report annotator agreement; given the subjectivity argument, report inter-annotator agreement (e.g., Cohen's kappa) and consider a larger or stratified sample.
[References] The reference list entry for Chapter 5 reads 'arXiv preprint arXiv:.' with no identifier; the full arXiv number should be supplied.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the thesis's central SPICED claims rest on held-out human annotations and external benchmarks, and the acknowledged 'papers as truth' assumption is a limitation, not a self-referential derivation.

full rationale

The thesis's main derivation chain is self-contained. SPICED is constructed by using SentenceBERT to propose candidate pairs and then human annotators to label the degree to which the paired findings describe the same scientific information ('the dataset ... is built by first pairing together potential scientific findings ... using SentenceBERT (SBERT), then presenting the potential pairs to human annotators', §1.4.3). The information matching score is therefore a human judgment, not a quantity derived from SBERT. The subsequent models are trained on those human labels and benchmarked against baselines ('I train several baseline models and benchmark their performance, finding that SBERT models fine-tuned on SPICED are best suited to the task', §1.4.3). The large-scale trend analyses apply the trained model to new unlabeled data, and the reported comparisons (general news vs. press releases, verified users vs. organizational accounts) depend on that model's generalization; any bias from SBERT-based candidate selection is a data-construction validity concern, not a case where the prediction reduces to its own input by construction. Chapter 7's zero-shot fact checking uses generated claims derived from the same corpus that supplies test evidence, which could raise dataset leakage concerns, but the veracity labels and the test claims are human-written and held out, so the result is not equivalent to the training input by construction. The thesis explicitly acknowledges the key assumption that 'scientific documents represent truth' (§1.5.3), and frames it as a limitation for normative conclusions rather than as a derived result. No load-bearing self-citation chain, no imported uniqueness theorem, and no fitted parameter renamed as a prediction were found. Overall, the derivation chain is not circular.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claims rest on human annotation as ground truth, SBERT-based candidate pre-selection, and the assumption that scientific texts are authoritative. There are no fitted physical constants; the free parameters are dataset curation choices and annotation constructs that shape the benchmark.

free parameters (2)

SentenceBERT threshold for SPICED candidate generation = not stated in excerpt
Pairs presented to annotators are pre-filtered by SentenceBERT similarity; this selection threshold determines which information-change pairs exist in the dataset and therefore shapes all downstream results.
Information Matching Score scale definitions = 1-5 ordinal scale
The IMS is an annotation construct introduced by the authors; the mapping of scale points to 'same finding' versus 'changed finding' is a design choice used as the training target for all SPICED models.

assumptions (4)

domain assumption Scientific documents are treated as the ground-truth source for what a finding actually says.
Stated in Section 1.5.3: 'I have assumed that scientific documents represent truth'; this assumption is load-bearing for interpreting information change as misinformation.
domain assumption Two sentences that paraphrase the same scientific finding can be identified by semantic similarity and human annotation.
The SPICED dataset and all downstream analyses rely on annotators agreeing on the Information Matching Score across papers, news, and Twitter, as described in Chapter 9.
domain assumption Statement-level comparison adequately captures the science communication pipeline.
The thesis models information change between individual sentences and acknowledges the 1-to-1 matching limitation in Section 1.5.3, which excludes multi-sentence and document-level transformations.
standard math Standard probabilistic and transformer-based NLP methods are valid tools for these tasks.
The thesis relies on BERT, RoBERTa, GPT-2, PU learning, and KL divergence as established methods; no new mathematical foundations are claimed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Machine Understanding of Scientific Language." pith.science (2026). https://pith.science/paper/WD45SZXX

@misc{pith2026250623990,
  author       = {Pith},
  title        = {Pith review of: Machine Understanding of Scientific Language},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WD45SZXX}},
  note         = {Machine review of arXiv:2506.23990}
}

read the original abstract

Scientific information expresses human understanding of nature. This knowledge is largely disseminated in different forms of text, including scientific papers, news articles, and discourse among people on social media. While important for accelerating our pursuit of knowledge, not all scientific text is faithful to the underlying science. As the volume of this text has burgeoned online in recent years, it has become a problem of societal importance to be able to identify the faithfulness of a given piece of scientific text automatically. This thesis is concerned with the cultivation of datasets, methods, and tools for machine understanding of scientific language, in order to analyze and understand science communication at scale. To arrive at this, I present several contributions in three areas of natural language processing and machine learning: automatic fact checking, learning with limited data, and scientific text processing. These contributions include new methods and resources for identifying check-worthy claims, adversarial claim generation, multi-source domain adaptation, learning from crowd-sourced labels, cite-worthiness detection, zero-shot scientific fact checking, detecting exaggerated scientific claims, and modeling degrees of information change in science communication. Critically, I demonstrate how the research outputs of this thesis are useful for effectively learning from limited amounts of scientific text in order to identify misinformative scientific statements and generate new insights into the science communication process

Figures

Figures reproduced from arXiv: 2506.23990 by the authors.

**Figure 1.1.** We are interested in measuring the information similarity of statements about [PITH_FULL_IMAGE:figures/full_fig_p021_1_1.png] view at source ↗

**Figure 1.** [PITH_FULL_IMAGE:figures/full_fig_p021_1.png] view at source ↗

**Figure 1.** [PITH_FULL_IMAGE:figures/full_fig_p026_1.png] view at source ↗

Figures from the paper (43 more)

**Figure 1.2.** Figure 1.2: In multi-source domain adaptation, a model is trained on data drawn from multiple [PITH_FULL_IMAGE:figures/full_fig_p027_1_2.png]

**Figure 1.3.** Figure 1.3: MT-PET design. We define pairs of complementary pattern-verbalizer pairs for a [PITH_FULL_IMAGE:figures/full_fig_p029_1_3.png]

**Figure 1.** Figure 1 [PITH_FULL_IMAGE:figures/full_fig_p029_1.png]

**Figure 2.1.** Figure 2.1: Examples of check-worthy and non check-worthy statements from three different [PITH_FULL_IMAGE:figures/full_fig_p037_2_1.png]

**Figure 2.** Figure 2 [PITH_FULL_IMAGE:figures/full_fig_p037_2.png]

**Figure 2.2.** Figure 2.2: High level view of PUC. A PU classifier (f, green box) is first learned using PU data (with s indicating if the sample is positive or unlabelled). From this the prior probability of a sample being positive is estimated. Unlabelled samples are then ranked by f (red bo…

**Figure 2.** Figure 2 [PITH_FULL_IMAGE:figures/full_fig_p041_2.png]

**Figure 3.1.** Figure 3.1: High level overview of our method. First, universal triggers are discovered for flipping [PITH_FULL_IMAGE:figures/full_fig_p050_3_1.png]

**Figure 4.1.** Figure 4.1: In multi-source domain adaptation, a model is trained on data drawn from multiple [PITH_FULL_IMAGE:figures/full_fig_p058_4_1.png]

**Figure 4.** Figure 4 [PITH_FULL_IMAGE:figures/full_fig_p058_4.png]

**Figure 4.2.** Figure 4.2: The overall approach tested in this work. A sample is input to a set of expert [PITH_FULL_IMAGE:figures/full_fig_p061_4_2.png]

**Figure 4.** Figure 4 [PITH_FULL_IMAGE:figures/full_fig_p067_4.png]

**Figure 4.3.** Figure 4.3: Final layer DistilBert embeddings for 500 randomly selected examples from each [PITH_FULL_IMAGE:figures/full_fig_p068_4_3.png]

**Figure 4.** Figure 4 [PITH_FULL_IMAGE:figures/full_fig_p068_4.png]

**Figure 4.4.** Figure 4.4: Comparison of agreement (Krippendorff’s alpha) between domain expert models [PITH_FULL_IMAGE:figures/full_fig_p069_4_4.png]

**Figure 5.1.** Figure 5.1: We experiment with four methods for aggregating soft labels in this work: Averaging, [PITH_FULL_IMAGE:figures/full_fig_p073_5_1.png]

**Figure 5.** Figure 5 [PITH_FULL_IMAGE:figures/full_fig_p080_5.png]

**Figure 5.2.** Figure 5.2: F1 scores for the (a) RTE, (b) MRE, (c) POS, and (d) Toxicity datasets on out of [PITH_FULL_IMAGE:figures/full_fig_p081_5_2.png]

**Figure 5.3.** Figure 5.3: Calibrated log-likelihood scores (lower is better) for the (a) RTE, (b) MRE, (c) POS, [PITH_FULL_IMAGE:figures/full_fig_p082_5_3.png]

**Figure 6.1.** Figure 6.1: Visualizing the BERT embeddings for 5 of the 10 domains from [PITH_FULL_IMAGE:figures/full_fig_p095_6_1.png]

**Figure 6.** Figure 6 [PITH_FULL_IMAGE:figures/full_fig_p095_6.png]

**Figure 7.1.** Figure 7.1: A complex excerpt from [159] (top) and the set of valid claims that can be generated from the bolded sentence (c1-c6). 7 Generating Scientific Claims for Zero-Shot Scientific Fact Checking 7.1 Introduction Scientific documents contain complex assertions about scienti…

**Figure 7.** Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p100_7.png]

**Figure 7.2.** Figure 7.2: KBIN method. We start with NER and linking to UMLS using scispaCy. We then find the most similar concepts with the same type using cui2vec, replace the entity in the source sentence using the canonical name and aliases of similar entities, and rank them using GPT-2. …

**Figure 7.** Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p104_7.png]

**Figure 7.** Figure 7 [PITH_FULL_IMAGE:figures/full_fig_p110_7.png]

**Figure 7.3.** Figure 7.3: Fact checking performance of models trained only on claims (i.e. no evidence). [PITH_FULL_IMAGE:figures/full_fig_p111_7_3.png]

**Figure 8.1.** Figure 8.1: Scientific exaggeration detection is the problem of identifying when a news article [PITH_FULL_IMAGE:figures/full_fig_p113_8_1.png]

**Figure 8.** Figure 8 [PITH_FULL_IMAGE:figures/full_fig_p115_8.png]

**Figure 8.2.** Figure 8.2: MT-PET design. We define pairs of complementary pattern-verbalizer pairs for a [PITH_FULL_IMAGE:figures/full_fig_p116_8_2.png]

**Figure 8.3.** Figure 8.3: Learning curve for supervised learning and PET compared to performance of [PITH_FULL_IMAGE:figures/full_fig_p124_8_3.png]

**Figure 8.4.** Figure 8.4: Proportion of examples by label which all models predict incorrectly. [PITH_FULL_IMAGE:figures/full_fig_p125_8_4.png]

**Figure 9.1.** Figure 9.1: We are interested in measuring the information similarity of statements about [PITH_FULL_IMAGE:figures/full_fig_p128_9_1.png]

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p128_9.png]

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p135_9.png]

**Figure 9.2.** Figure 9.2: (a) Mean Squared Error (MSE, ↓ better) and (b) Pearson correlation (r, ↑ better) on the test set of SPICED. Grey = zero-shot transfer models, red = MLM models fine-tuned on SPICED, blue = SBERT models fine-tuned on SPICED. Results are averaged across 5 random seeds. …

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p136_9.png]

**Figure 9.3.** Figure 9.3: Scientific findings covered by Press Release and SciTech generally have less [PITH_FULL_IMAGE:figures/full_fig_p139_9_3.png]

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p139_9.png]

**Figure 9.4.** Figure 9.4: Organizational Twitter accounts keep more original information from the paper finding [PITH_FULL_IMAGE:figures/full_fig_p140_9_4.png]

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p140_9.png]

**Figure 9.5.** Figure 9.5: Journalists tend to downplay the certainty and strength of findings in abstracts, but [PITH_FULL_IMAGE:figures/full_fig_p141_9_5.png]

**Figure 9.** Figure 9 [PITH_FULL_IMAGE:figures/full_fig_p142_9.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

286 extracted references · 60 canonical work pages

[1]

A Graph-Theoretic Embedding- Based Approach for Rumor Detection in Twitter

ABULAISH , M., K UMARI , N., F AZIL , M., AND SINGH , B. A Graph-Theoretic Embedding- Based Approach for Rumor Detection in Twitter. In IEEE/WIC/ACM International Conference on Web Intelligence (2019), pp. 466–470

2019
[2]

Unsupervised Domain Clusters in Pretrained Language Models

AHARONI , R., AND GOLDBERG , Y. Unsupervised Domain Clusters in Pretrained Language Models. In ACL (2020)

2020
[3]

Time-Aware Evidence Ranking for Fact- Checking

ALLEIN , L., A UGENSTEIN , I., AND MOENS , M.-F. Time-Aware Evidence Ranking for Fact- Checking. arXiv preprint arXiv:2009.06402 (2020)

arXiv 2020
[4]

Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning

ALLEN -ZHU, Z., AND LI, Y. Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning. CoRR abs/2012.09816 (2020)

arXiv 2020
[5]

S., T HORNE , J., V LACHOS , A., C HRISTODOULOPOU - LOS, C., C OCARASCU , O., AND MITTAL , A

ALY, R., G UO, Z., S CHLICHTKRULL , M. S., T HORNE , J., V LACHOS , A., C HRISTODOULOPOU - LOS, C., C OCARASCU , O., AND MITTAL , A. FEVEROUS: fact extraction and verification over unstructured and structured information. In Proceedings of the Neural Information Pro- cessing Systems Track on Datasets and Benchmarks 1, NeurIPS Datasets and Benchmarks 2021,...

2021
[6]

Construction of the Literature Graph in Semantic Scholar

AMMAR , W., G ROENEVELD , D., B HAGAVATULA , C., B ELTAGY, I., C RAWFORD , M., D OWNEY, D., D UNKELBERGER , J., E LGOHARY , A., F ELDMAN , S., H A, V., ET AL . Construction of the Literature Graph in Semantic Scholar. NAACL HLT 2018 (2018), 84–91

2018
[7]

ASHUKHA , A., L YZHOV, A., M OLCHANOV , D., AND VETROV, D. P. Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020 (2020), OpenReview.net

2020
[8]

ATANASOVA , P., B ARRON -CEDENO , A., E LSAYED , T., S UWAILEH , R., Z AGHOUANI , W., KYUCHUKOV , S., M ARTINO , G. D. S., AND NAKOV, P .Overview of the CLEF-2018 CheckThat! Lab on Automatic Identification and Verification of Political Claims. Task 1: Check-Worthiness. arXiv preprint arXiv:1808.05542 (2018)

arXiv 2018

Show all 286 references

[9]

G., L IOMA , C., AND AUGENSTEIN , I

ATANASOVA , P., S IMONSEN , J. G., L IOMA , C., AND AUGENSTEIN , I. A Diagnostic Study of Explainability Techniques for Text Classification. In Proceedings of EMNLP (2020), Association for Computational Linguistics

2020
[10]

G., L IOMA , C., AND AUGENSTEIN , I

ATANASOVA , P., S IMONSEN , J. G., L IOMA , C., AND AUGENSTEIN , I. Generating fact checking explanations. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020, Online, July 5-10, 2020 (2020), D. Jurafsky, J. Chai, 132 N. Schlute...

2020
[11]

Generating Label Cohesive and Well- Formed Adversarial Claims

ATANASOVA , P., WRIGHT , D., AND AUGENSTEIN , I. Generating Label Cohesive and Well- Formed Adversarial Claims. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) (Online, Nov. 2020), Association for Computational Linguistics, pp....

2020
[12]

SemEval 2017 Task 10: ScienceIE-Extracting Keyphrases and Relations from Scientific Publications

AUGENSTEIN , I., D AS, M., R IEDEL , S., V IKRAMAN , L., AND MCCALLUM , A. SemEval 2017 Task 10: ScienceIE-Extracting Keyphrases and Relations from Scientific Publications. In Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017) (2017), pp. 546–555

2017
[13]

AUGENSTEIN , I., L IOMA , C., W ANG , D., C HAVES LIMA , L., H ANSEN , C., H ANSEN , C., AND SIMONSEN , J. G. MultiFC: A real-world multi-domain dataset for evidence-based fact checking of claims. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language P...

2019
[14]

Stance Detection with Bidirectional Conditional Encoding

AUGENSTEIN , I., R OCKTÄSCHEL , T., V LACHOS , A., AND BONTCHEVA , K. Stance Detection with Bidirectional Conditional Encoding. In EMNLP (2016), J. Su, X. Carreras, and K. Duh, Eds., The Association for Computational Linguistics, pp. 876–885

2016
[15]

Multi-Task Learning of Keyphrase Boundary Classi- fication

AUGENSTEIN , I., AND SØGAARD , A. Multi-Task Learning of Keyphrase Boundary Classi- fication. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (Vancouver, Canada, July 2017), Association for Computational Lingu...

2017
[16]

A., AND REINECKE , K

AUGUST , T., C ARD, D., H SIEH , G., S MITH , N. A., AND REINECKE , K. Explain like I am a scientist: The linguistic barriers of entry to r/science. In CHI ’20: CHI Conference on Human Factors in Computing Systems, Honolulu, HI, USA, April 25-30, 2020 (2020), R. Bernhaupt, F ....

2020
[17]

AUGUST , T., K IM, L., R EINECKE , K., AND SMITH , N. A. Writing strategies for science communication: Data and computational analysis. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) (Online, 2020), Association for Computation...

2020
[18]

BA, J., AND CARUANA , R. Do Deep Nets Really Need to be Deep? In Advances in Neural Information Processing Systems 27: Annual Conference on Neural Information Processing 133 Systems 2014, December 8-13 2014, Montreal, Quebec, Canada (2014), Z. Ghahramani, M. Welling, C. Cortes...

2014
[19]

D., W RIGHT , D., K ATSIS, Y., KIM, H.-C., S WAFFORD , A

BADAL , V. D., W RIGHT , D., K ATSIS, Y., KIM, H.-C., S WAFFORD , A. D., K NIGHT , R., AND HSU, C.-N. Challenges in the construction of knowledge bases for human microbiome- disease associations. Microbiome 7, 1 (2019), 1–15

2019
[20]

M., M ASON , W

BAKSHY, E., H OFMAN , J. M., M ASON , W. A., AND WATTS, D. J. Everyone’s an influencer: quantifying influence on twitter. In Proceedings of the Forth International Conference on Web Search and Web Data Mining, WSDM 2011, Hong Kong, China, February 9-12, 2011 (2011), I. King, W...

2011
[21]

R., M ÀRQUEZ , L., M OSCHITTI , A., AND NAKOV, P

BALY, R., M OHTARAMI , M., G LASS , J. R., M ÀRQUEZ , L., M OSCHITTI , A., AND NAKOV, P. Integrating Stance Detection and Fact Checking in a Unified Corpus. In NAACL-HLT (2) (2018), M. A. Walker, H. Ji, and A. Stent, Eds., Association for Computational Linguistics, pp. 21–27

2018
[22]

CheckThat! at CLEF 2020: Enabling the Automatic Identification and Verification of Claims in Social Media

BARRÓN -CEDEÑO , A., E LSAYED , T., N AKOV, P., DA SAN MARTINO , G., H ASANAIN , M., SUWAILEH , R., AND HAOUARI , F. CheckThat! at CLEF 2020: Enabling the Automatic Identification and Verification of Claims in Social Media. In European Conference on Information Retrieval (2020...

2020
[23]

Overview of the CLEF-2018 CheckThat! Lab on Automatic Identification and Verification of Political Claims

BARRÓN -CEDEÑO , A., E LSAYED , T., S UWAILEH , R., M ÀRQUEZ , L., A TANASOVA , P., ZAGHOUANI , W., K YUCHUKOV , S., D A SAN MARTINO , G., AND NAKOV, P. Overview of the CLEF-2018 CheckThat! Lab on Automatic Identification and Verification of Political Claims. Task 2: Factuality

2018
[24]

L., K OMPA, B., S CHMALTZ , A., F RIED, I., W EBER , G

BEAM , A. L., K OMPA, B., S CHMALTZ , A., F RIED, I., W EBER , G. M., P ALMER , N. P., SHI, X., C AI, T., AND KOHANE , I. S. Clinical Concept Embeddings Learned from Massive Sources of Multimodal Medical Data. In Pacific Symposium on Biocomputing 2020, Fairmont Orchid, Hawaii,...

2020
[25]

Learning from positive and unlabeled data: A survey

BEKKER , J., AND DAVIS, J. Learning from positive and unlabeled data: A survey. arXiv preprint arXiv:1811.04820 (2018)

2018 arXiv
[26]

SciBERT: A Pretrained Language Model for Scientific Text

BELTAGY, I., L O, K., AND COHAN , A. SciBERT: A Pretrained Language Model for Scientific Text. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) (20...

2019
[27]

E., AND COHAN , A

BELTAGY, I., P ETERS , M. E., AND COHAN , A. Longformer: The Long-Document Trans- former. CoRR abs/2004.05150 (2020). 134

2020 arXiv
[28]

NLTK: The Natural Language Toolkit

BIRD, S. NLTK: The Natural Language Toolkit. In ACL 2006, 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, Sydney, Australia, 17-21 July 2006 (2006), N. Calzolari...

2006
[29]

J., G IBSON , B., J OSEPH , M

BIRD, S., D ALE , R., D ORR , B. J., G IBSON , B., J OSEPH , M. T., K AN, M.-Y., L EE, D., POWLEY, B., R ADEV, D. R., AND TAN, Y. F. The ACL Anthology Reference Corpus: A Reference Dataset for Bibliographic Research in Computational Linguistics
[30]

Biographies, Bollywood, Boom-Boxes and Blenders: Domain Adaptation for Sentiment Classification

BLITZER , J., D REDZE , M., AND PEREIRA , F. Biographies, Bollywood, Boom-Boxes and Blenders: Domain Adaptation for Sentiment Classification. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics (2007), pp. 440–447

2007
[31]

Domain Adaptation with Structural Cor- respondence Learning

BLITZER , J., M CDONALD , R., AND PEREIRA , F. Domain Adaptation with Structural Cor- respondence Learning. In Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing (Sydney, Australia, July 2006), Association for Computational Linguistics, pp. 120–128

2006
[32]

BODE , L., AND VRAGA , E. K. See something, say something: Correction of global health misinformation on social media. Health communication 33, 9 (2018), 1131–1140

2018
[33]

The Unified Medical Language System (UMLS): integrating biomedical terminology

BODENREIDER , O. The Unified Medical Language System (UMLS): integrating biomedical terminology. Nucleic Acids Res. 32, Database-Issue (2004), 267–270

2004
[34]

Explainable assess- ment of healthcare articles with QA

BOISSONNET , A., S AEIDI , M., P LACHOURAS , V., AND VLACHOS , A. Explainable assess- ment of healthcare articles with QA. In Proceedings of the 21st Workshop on Biomedical Language Processing, BioNLP@ACL 2022, Dublin, Ireland, May 26, 2022(2022), D. Demner- Fushman, K. B. Coh...

2022
[35]

Enriching Word Vectors with Subword Information

BOJANOWSKI , P., GRAVE , E., J OULIN , A., AND MIKOLOV, T. Enriching Word Vectors with Subword Information. Transactions of the Association for Computational Linguistics 5(2017), 135–146

2017
[36]

Inducing Relational Knowledge from BERT

BOURAOUI , Z., C AMACHO -COLLADOS , J., AND SCHOCKAERT , S. Inducing Relational Knowledge from BERT. In The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, The Thirty-Second Innovative Applications of Artificial Intelligence Conference, IAAI 2020, The Tent...

2020
[37]

R., A NGELI , G., P OTTS , C., AND MANNING , C

BOWMAN , S. R., A NGELI , G., P OTTS , C., AND MANNING , C. D. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical 135 Methods in Natural Language Processing, EMNLP 2015, Lisbon, Portugal, September 17-21, 2015 (...

2015
[38]

C., C HALLENGER , A., B OIVIN , J., B OTT, L., C HAMBERS , C

BRATTON , L., A DAMS , R. C., C HALLENGER , A., B OIVIN , J., B OTT, L., C HAMBERS , C. D., AND SUMNER , P. The Association Between Exaggeration in Health-Related Science News and Academic Press Releases: A Replication Study. Wellcome open research 4 (2019)

2019
[39]

BROWN , T. B., M ANN , B., R YDER , N., S UBBIAH , M., K APLAN , J., D HARIWAL , P., N EE- LAKANTAN , A., S HYAM, P., SASTRY, G., A SKELL , A., A GARWAL , S., H ERBERT -VOSS, A., KRUEGER , G., H ENIGHAN , T., C HILD, R., R AMESH , A., Z IEGLER , D. M., W U, J., W INTER , C., H...

2020
[40]

S., I OANNIDIS , J., M OKRYSZ , C., N OSEK , B

BUTTON , K. S., I OANNIDIS , J., M OKRYSZ , C., N OSEK , B. A., F LINT, J., R OBINSON , E. S., AND MUNAFÒ , M. R. Power failure: why small sample size undermines the reliability of neuroscience. Nature reviews neuroscience 14, 5 (2013), 365–376

2013
[41]

YAKE! Keyword extraction from single documents using multiple local features.Inf

CAMPOS , R., M ANGARAVITE , V., PASQUALI , A., J ORGE , A., N UNES , C., AND JATOWT, A. YAKE! Keyword extraction from single documents using multiple local features.Inf. Sci. 509 (2020), 257–289

2020
[42]

Pubmed: the bibliographic database

CANESE , K., AND WEIS, S. Pubmed: the bibliographic database. The NCBI handbook 2, 1 (2013)

2013
[43]

Multilevel Bayesian Models of Categorical Data Annotation

CARPENTER , B. Multilevel Bayesian Models of Categorical Data Annotation. Unpublished manuscript 17, 122 (2008), 45–50

2008
[44]

SemEval-2017 task 1: Semantic textual similarity multilingual and crosslingual focused evaluation

CER, D., D IAB, M., A GIRRE , E., L OPEZ -GAZPIO , I., AND SPECIA , L. SemEval-2017 task 1: Semantic textual similarity multilingual and crosslingual focused evaluation. In Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017) (Vancouver, Canada,...

2017
[45]

K., F EIGENBLAT , G., H OVY, E

CHANDRASEKARAN , M. K., F EIGENBLAT , G., H OVY, E. H., R AVICHANDER , A., S HMUELI - SCHEUER , M., AND DE WAARD, A. Overview and insights from the shared tasks at scholarly document processing 2020: Cl-scisumm, laysumm and longsumm. In Proceedings of the First Workshop on Sch...

2020
[46]

Decon- textualization: Making Sentences Stand-Alone

CHOI , E., P ALOMAKI , J., L AMM , M., K WIATKOWSKI , T., DAS, D., AND COLLINS , M. Decon- textualization: Making Sentences Stand-Alone. Trans. Assoc. Comput. Linguistics 9 (2021), 447–461

2021
[47]

C LARK , S. K. Writing strategies for science. Teacher Created Materials, 2013

2013
[48]

Structural Scaffolds for Citation Intent Classification in Scientific Publications

C OHAN , A., A MMAR , W., VAN ZUYLEN , M., AND CADY, F. Structural Scaffolds for Citation Intent Classification in Scientific Publications. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Tech...

2019
[49]

SPECTER: Document- level representation learning using citation-informed transformers

COHAN , A., F ELDMAN , S., B ELTAGY, I., D OWNEY, D., AND WELD, D. SPECTER: Document- level representation learning using citation-informed transformers. InProceedings of the 58th Annual Meeting of the Association for Computational Linguistics (Online, 2020), Association for C...

2020
[50]

A supervised approach to extractive summarisation of scientific papers

COLLINS , E., A UGENSTEIN , I., AND RIEDEL , S. A supervised approach to extractive summarisation of scientific papers. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017), Vancouver, Canada, August 3-4, 2017 (2017), R. Levy and L. Spe...

2017
[51]

Science reporting to the public: Does the message get twisted? CMAJ 170, 9 (2004), 1415–1416

CONDIT, C. Science reporting to the public: Does the message get twisted? CMAJ 170, 9 (2004), 1415–1416

2004
[52]

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

CONNEAU , A., K IELA , D., S CHWENK , H., B ARRAULT, L., AND BORDES , A. Supervised Learning of Universal Sentence Representations from Natural Language Inference Data. In EMNLP 2017 (2017), pp. 670–680

2017
[53]

The PASCAL Recognising Textual Entailment Challenge

DAGAN , I., G LICKMAN , O., AND MAGNINI , B. The PASCAL Recognising Textual Entailment Challenge. In Machine Learning Challenges, Evaluating Predictive Uncertainty, Visual Object Classification and Recognizing Textual Entailment, First PASCAL Machine Learning Challenges Worksh...

2005
[54]

Ginger cannot cure cancer: Battling fake health news with a comprehensive data repository

DAI, E., S UN, Y., AND WANG , S. Ginger cannot cure cancer: Battling fake health news with a comprehensive data repository. In Proceedings of the Fourteenth International AAAI Conference on Web and Social Media, ICWSM 2020, Held Virtually, Original Venue: Atlanta, 137 Georgia,...

2020
[55]

We Can Explain Y our Research in Layman’s Terms: Towards Automating Science Journalism at Scale

DANGOVSKI , R., S HEN , M., B YRD, D., J ING , L., T SVETKOVA , D., N AKOVA, P., AND SOLJACIC , M. We Can Explain Y our Research in Layman’s Terms: Towards Automating Science Journalism at Scale. In AAAI 2021 (2021), AAAI Press

2021
[56]

Frustratingly Easy Domain Adaptation

DAUMÉ , III, H. Frustratingly Easy Domain Adaptation. In Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics (Prague, Czech Republic, June 2007), Association for Computational Linguistics, pp. 256–263

2007
[57]

P ., AND SKENE , A

DAWID, A. P ., AND SKENE , A. M. Maximum Likelihood Estimation of Observer Error-Rates Using the EM Algorithm. Journal of the Royal Statistical Society: Series C (Applied Statistics) 28, 1 (1979), 20–28

1979
[58]

Positive and Unlabeled Examples Help Learning

DE COMITÉ , F., D ENIS , F., G ILLERON , R., AND LETOUZEY , F. Positive and Unlabeled Examples Help Learning. In International Conference on Algorithmic Learning Theory (1999), Springer, pp. 219–230

1999
[59]

E., AND QUATTROCIOCCHI , W

DEL VICARIO , M., B ESSI , A., Z OLLO , F., P ETRONI , F., S CALA , A., C ALDARELLI , G., STANLEY, H. E., AND QUATTROCIOCCHI , W. The Spreading of Misinformation Online. Proceedings of the National Academy of Sciences 113, 3 (2016), 554–559

2016
[60]

Transforming Question Answering Datasets Into Natural Language Inference Datasets

DEMSZKY , D., G UU, K., AND LIANG , P. Transforming Question Answering Datasets Into Natural Language Inference Datasets. CoRR abs/1809.02922 (2018)

2018 arXiv
[61]

PAC Learning From Positive Statistical Queries

DENIS , F. PAC Learning From Positive Statistical Queries. In International Conference on Algorithmic Learning Theory (1998), Springer, pp. 112–126

1998
[62]

BERT: Pre-Training of Deep Bidi- rectional Transformers for Language Understanding

DEVLIN , J., C HANG , M.-W., L EE, K., AND TOUTANOVA , K. BERT: Pre-Training of Deep Bidi- rectional Transformers for Language Understanding. In NAACL-HLT 2019 (2019), pp. 4171– 4186

2019
[63]

DEYOUNG , J., B ELTAGY, I., VAN ZUYLEN , M., K UEHL , B., AND WANG , L. L. MS2: Multi- Document Summarization of Medical Studies. CoRR abs/2104.06486 (2021)

2021 arXiv
[64]

A., B ARSHAN , E., AND BABANEZHAD , R

DIA, O. A., B ARSHAN , E., AND BABANEZHAD , R. Semantics Preserving Adversarial Learning. arXiv preprint arXiv:1903.03905 (2019)

2019 arXiv
[65]

I., L EAMAN , R., AND LU, Z

DO ˘GAN , R. I., L EAMAN , R., AND LU, Z. NCBI Disease Corpus: a Resource for Disease Name Recognition and Concept Normalization. Journal of Biomedical Informatics 47 (2014), 1–10. 138

2014
[66]

Semi-supervised Domain Adaptation with Instance Constraints

DONAHUE , J., H OFFMAN , J., R ODNER , E., S AENKO , K., AND DARRELL , T. Semi-supervised Domain Adaptation with Instance Constraints. In CVPR (2013), IEEE Computer Society, pp. 668–675

2013
[67]

C., N IU, G., AND SUGIYAMA , M

DU PLESSIS , M. C., N IU, G., AND SUGIYAMA , M. Analysis of Learning From Positive and Unlabeled Data. In Advances in Neural Information Processing Systems (2014), pp. 703– 711

2014
[68]

Question Generation for Question Answer- ing

DUAN, N., T ANG , D., C HEN , P., AND ZHOU, M. Question Generation for Question Answer- ing. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017, Copenhagen, Denmark, September 9-11, 2017(2017), M. Palmer, R. Hwa, and S. Riedel...

2017
[69]

Crowdsourcing Semantic Label Propagation in Relation Classification

DUMITRACHE , A., A ROYO, L., AND WELTY, C. Crowdsourcing Semantic Label Propagation in Relation Classification. CoRR abs/1809.00537 (2018)

2018 arXiv
[70]

HotFlip: White-Box Adversarial Examples for Text Classification

EBRAHIMI , J., R AO, A., L OWD, D., AND DOU, D. HotFlip: White-Box Adversarial Examples for Text Classification. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (2018), pp. 31–36

2018
[71]

Learning Classifiers From Only Positive and Unlabeled Data

ELKAN , C., AND NOTO, K. Learning Classifiers From Only Positive and Unlabeled Data. In Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Minin (2008), pp. 213–220

2008
[72]

Overview of the CLEF-2019 CheckThat! Lab: Automatic Identification and Verification of Claims

ELSAYED , T., N AKOV, P., B ARRÓN -CEDEÑO , A., H ASANAIN , M., S UWAILEH , R., DA SAN MARTINO , G., AND ATANASOVA , P. Overview of the CLEF-2019 CheckThat! Lab: Automatic Identification and Verification of Claims. In International Conference of the Cross-Language Evaluation F...

2019
[73]

U., P ING , Z., L I, Y., XU, Y., ET AL

FANG , X., W ANG , K., H AN, D., H E, X., W EI, J., Z HAO, L., I MAM , M. U., P ING , Z., L I, Y., XU, Y., ET AL . Dietary magnesium intake and the risk of cardiovascular disease, type 2 diabetes, and all-cause mortality: a dose–response meta-analysis of prospective cohort stu...

2016
[74]

A High-Quality Gold Standard for Citation- Based Tasks

FÄRBER , M., T HIEMANN , A., AND JATOWT, A. A High-Quality Gold Standard for Citation- Based Tasks. In Proceedings of the Eleventh International Conference on Language Re- sources and Evaluation (LREC 2018) (2018)

2018
[75]

To Cite, or Not to Cite? Detecting Citation Contexts in Text

FÄRBER , M., T HIEMANN , A., AND JATOWT, A. To Cite, or Not to Cite? Detecting Citation Contexts in Text. InEuropean Conference on Information Retrieval(2018), Springer, pp. 598– 603. 139

2018
[76]

R., AND MANNING , C

FINKEL , J. R., AND MANNING , C. D. Hierarchical Bayesian Domain Adaptation. In Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics (Boulder, Colorado, June 2009), Association for...

2009
[77]

Communicating uncertainty: Fulfilling the duty to inform

FISCHHOFF , B. Communicating uncertainty: Fulfilling the duty to inform. Issues in Science and Technology 28 (2012)

2012
[78]

Beyond Black & White: Leveraging Annotator Disagreement via Soft-Label Multi-Task Learning

FORNACIARI , T., U MA, A., P AUN, S., P LANK , B., H OVY, D., AND POESIO , M. Beyond Black & White: Leveraging Annotator Disagreement via Soft-Label Multi-Task Learning. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Li...

2021
[79]

Linear mixed-effects model

G AŁECKI , A., AND BURZYKOWSKI , T. Linear mixed-effects model. In Linear mixed-effects models using R. Springer, 2013, pp. 245–273

2013
[80]

Unsupervised Domain Adaptation by Backpropagation

GANIN , Y., AND LEMPITSKY , V. Unsupervised Domain Adaptation by Backpropagation. In International Conference on Machine Learning (2015), pp. 1180–1189

2015
[81]

PPDB: The paraphrase database

GANITKEVITCH , J., V AN DURME , B., AND CALLISON -BURCH , C. PPDB: The paraphrase database. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies(Atlanta, Georgia, 2013), Association fo...

2013
[82]

Universal Adversarial Perturbation for Text Classification.arXiv preprint arXiv:1910.04618 (2019)

GAO, H., AND OATES, T. Universal Adversarial Perturbation for Text Classification.arXiv preprint arXiv:1910.04618 (2019)

2019 arXiv
[83]

A Context-Aware Approach for Detecting Worth-Checking Claims in Political Debates

GENCHEVA , P., N AKOV, P., M ÀRQUEZ , L., B ARRÓN -CEDEÑO , A., AND KOYCHEV, I. A Context-Aware Approach for Detecting Worth-Checking Claims in Political Debates. In Pro- ceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017 (Varna...

2017
[84]

I., L IU, V., H UANG , S., L EE, J., AND CHILTON , L

GERO, K. I., L IU, V., H UANG , S., L EE, J., AND CHILTON , L. B. What makes tweetorials tick: How experts communicate complex topics on twitter. Proceedings of the ACM on Human-Computer Interaction 5, CSCW2 (2021), 1–26

2021
[85]

GIMPEL , K., S CHNEIDER , N., O’C ONNOR , B., D AS, D., M ILLS , D., E ISENSTEIN , J., HEILMAN , M., Y OGATAMA , D., F LANIGAN , J., AND SMITH , N. A. Part-of-Speech Tagging for Twitter: Annotation, Features, and Experiments. In The 49th Annual Meeting of the 140 Association f...

2011
[86]

GINEV, D., AND MILLER , B. R. Scientific statement classification over arXiv.org. In Proceedings of the 12th Language Resources and Evaluation Conference(Marseille, France, 2020), European Language Resources Association, pp. 1219–1226

2020
[87]

J., S HLENS , J., AND SZEGEDY, C

GOODFELLOW , I. J., S HLENS , J., AND SZEGEDY, C. Explaining and Harnessing Adversarial Examples. stat 1050 (2015), 20

2015
[88]

L., Z HOU, K., P ATEL, K., H ASHIMOTO , T., AND BERNSTEIN , M

GORDON , M. L., Z HOU, K., P ATEL, K., H ASHIMOTO , T., AND BERNSTEIN , M. S. The Disagreement Deconvolution: Bringing Machine Learning Performance Metrics In Line With Reality. In CHI ’21: CHI Conference on Human Factors in Computing Systems, Virtual Event / Y okohama, Japan,...

2021
[89]

Is Y our Toxicity My Toxicity? Exploring the Impact of Rater Identity on Toxicity Annotation

GOYAL, N., K IVLICHAN , I., R OSEN , R., AND VASSERMAN , L. Is Y our Toxicity My Toxicity? Exploring the Impact of Rater Identity on Toxicity Annotation. CoRR abs/2205.00501 (2022)

2022 arXiv
[90]

The Rise of Fact-Checking Sites in Europe

GRAVES , L., AND CHERUBINI , F. The Rise of Fact-Checking Sites in Europe. Reuters Institute for the Study of Journalism (2016)

2016
[91]

Part-of-Speech Tagging for Twitter with Adversarial Neural Networks

GUI, T., Z HANG , Q., H UANG , H., P ENG , M., AND HUANG , X.-J. Part-of-Speech Tagging for Twitter with Adversarial Neural Networks. In EMNLP 2017 (2017), pp. 2411–2420

2017
[92]

Multi-Source Domain Adaptation with Mixture of Experts

GUO, J., S HAH , D., AND BARZILAY , R. Multi-Source Domain Adaptation with Mixture of Experts. In EMNLP 2018 (2018), pp. 4694–4703

2018
[93]

S., AND VLACHOS , A

GUO, Z., S CHLICHTKRULL , M. S., AND VLACHOS , A. A survey on automated fact-checking. Trans. Assoc. Comput. Linguistics 10 (2022), 178–206

2022
[94]

GURURANGAN , S., M ARASOVIC , A., S WAYAMDIPTA , S., L O, K., B ELTAGY, I., D OWNEY, D., AND SMITH , N. A. Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020, ...

2020
[95]

GUSTAFSON , A., AND RICE , R. E. The effects of uncertainty frames in three science communication topics. Science Communication 41, 6 (2019), 679–706

2019
[96]

Unsupervised Domain Adaptation of Contextualized Embed- dings: A Case Study in Early Modern English

HAN, X., AND EISENSTEIN , J. Unsupervised Domain Adaptation of Contextualized Embed- dings: A Case Study in Early Modern English. 4229–4239. 141
[97]

Neural Check-Worthiness Ranking With Weak Supervision: Finding Sentences for Fact-Checking

HANSEN , C., H ANSEN , C., A LSTRUP, S., G RUE SIMONSEN , J., AND LIOMA , C. Neural Check-Worthiness Ranking With Weak Supervision: Finding Sentences for Fact-Checking. In Companion Proceedings of the 2019 World Wide Web Conference (2019), pp. 994–1000

2019
[98]

A Survey on Stance Detection for Mis- and Disinformation Identification, 2021

HARDALOV , M., A RORA , A., N AKOV, P .,AND AUGENSTEIN , I. A Survey on Stance Detection for Mis- and Disinformation Identification, 2021

2021
[99]

Cross-Domain Label- Adaptive Stance Detection

HARDALOV , M., A RORA , A., N AKOV, P., AND AUGENSTEIN , I. Cross-Domain Label- Adaptive Stance Detection. In Proceedings of EMNLP (2021), Association for Computational Linguistics

2021
[100]

Few-Shot Cross-Lingual Stance Detection with Sentiment-Based Pre-Training

HARDALOV , M., A RORA , A., N AKOV, P., AND AUGENSTEIN , I. Few-Shot Cross-Lingual Stance Detection with Sentiment-Based Pre-Training. CoRR (2021)

2021
[101]

S., AND FELDMAN , L

HART, P. S., AND FELDMAN , L. The impact of climate change–related imagery and text on public opinion and behavior change. Science Communication 38, 4 (2016), 415–441

2016
[102]

K., ET AL

HASSAN , N., Z HANG , G., A RSLAN , F., C ARABALLO , J., J IMENEZ , D., G AWSANE , S., HASAN , S., J OSEPH , M., K ULKARNI , A., N AYAK, A. K., ET AL . ClaimBuster: the First-Ever End-to-End Fact-Checking System. Proceedings of the VLDB Endowment 10, 12 (2017), 1945–1948

2017
[103]

Finding users we trust: Scaling up verified twitter users using their communication patterns

HENTSCHEL , M., A LONSO , O., C OUNTS , S., AND KANDYLAS , V. Finding users we trust: Scaling up verified twitter users using their communication patterns. In Eighth International AAAI Conference on Weblogs and Social Media (2014)

2014
[104]

DeSePtion: Dual sequence prediction and adversarial examples for im- proved fact-checking

HIDEY, C., C HAKRABARTY , T., A LHINDI , T., VARIA , S., K RSTOVSKI , K., D IAB, M., AND MURESAN , S. DeSePtion: Dual sequence prediction and adversarial examples for im- proved fact-checking. In Proceedings of the 58th Annual Meeting of the Association for Computational Lingu...

2020
[105]

E., V INYALS , O., AND DEAN , J

HINTON , G. E., V INYALS , O., AND DEAN , J. Distilling the knowledge in a neural network. CoRR abs/1503.02531 (2015)

2015 arXiv
[106]

Long short-term memory

HOCHREITER , S., AND SCHMIDHUBER , J. Long short-term memory. Neural computation 9, 8 (1997), 1735–1780

1997
[107]

N., P LANK , B., W RIGHT , D., AND AUGENSTEIN , I

HOLM , A. N., P LANK , B., W RIGHT , D., AND AUGENSTEIN , I. Longitudinal Citation Prediction using Temporal Graph Neural Networks. arXiv preprint arXiv:2012.05742 (2020)

2020 arXiv
[108]

HOLUB , A., P ERONA , P., AND BURL , M. C. Entropy-based active learning for object recognition. In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 142 Workshops 2008, Anchorage, AK, USA, 23-28 June, 2008 (2008), IEEE Computer Society, pp. 1–8

2008
[109]

Bayesian active learning for classification and preference learning

HOULSBY , N., H USZAR , F., GHAHRAMANI , Z., AND LENGYEL , M. Bayesian active learning for classification and preference learning. CoRR abs/1112.5745 (2011)

2011 arXiv
[110]

Learning whom to trust with MACE

HOVY, D., B ERG -KIRKPATRICK , T., VASWANI , A., AND HOVY, E. Learning whom to trust with MACE. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies(Atlanta, Georgia, 2013), Associatio...

2013
[111]

Experiments with crowdsourced re-annotation of a POS tagging data set

HOVY, D., P LANK , B., AND SØGAARD , A. Experiments with crowdsourced re-annotation of a POS tagging data set. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014, June 22-27, 2014, Baltimore, MD, USA, Volume 2: Short Papers (20...

2014
[112]

Universal language model fine-tuning for text classification

HOWARD, J., AND RUDER , S. Universal language model fine-tuning for text classification. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguis- tics, ACL 2018, Melbourne, Australia, July 15-20, 2018, Volume 1: Long Papers (2018), I. Gurevych an...

2018
[113]

Digital Wildfires in a Hyperconnected World

HOWELL , L., ET AL . Digital Wildfires in a Hyperconnected World. WEF report 3 (2013), 15–94

2013
[114]

IOANNIDIS , J. P. Why most published research findings are false. PLoS medicine 2, 8 (2005), e124

2005
[115]

JARADAT, I., G ENCHEVA , P ., BARRÓN -CEDEÑO , A., M ÀRQUEZ , L., AND NAKOV, P .Claim- rank: Detecting Check-Worthy Claims in Arabic and English. 26–30
[116]

F., A RAKI , J., AND NEUBIG , G

JIANG , Z., X U, F. F., A RAKI , J., AND NEUBIG , G. How can we know what language models know. Trans. Assoc. Comput. Linguistics 8 (2020), 423–438

2020
[117]

T., AND SZOLOVITS , P

JIN, D., J IN, Z., Z HOU, J. T., AND SZOLOVITS , P. TextFool: Fool your Model with Natural Adversarial Text
[118]

Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing

JOLLY, S., A TANASOVA , P., AND AUGENSTEIN , I. Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing. Information 13 (2022)

2022
[119]

Measuring the Evolution of a Scientific Field through Citation Frames

JÜRGENS , D., K UMAR , S., H OOVER , R., M CFARLAND , D., AND JURAFSKY , D. Measuring the Evolution of a Scientific Field through Citation Frames. Transactions of the Association for Computational Linguistics 6 (2018), 391–406. 143

2018
[120]

S., M CCANN , B., V ARSHNEY , L

KESKAR , N. S., M CCANN , B., V ARSHNEY , L. R., X IONG , C., AND SOCHER , R. Ctrl: A conditional transformer language model for controllable generation. arXiv preprint arXiv:1909.05858 (2019)

2019 arXiv
[121]

GENIA Corpus – a Semantically Annotated Corpus for Bio-Textmining

KIM, J.-D., O HTA, T., T ATEISI , Y., AND TSUJII , J. GENIA Corpus – a Semantically Annotated Corpus for Bio-Textmining. Bioinformatics 19, suppl_1 (2003), i180–i182

2003
[122]

Introduction to the Bio-Entity Recognition Task at JNLPBA

KIM, J.-D., O HTA, T., T SURUOKA , Y., TATEISI , Y., AND COLLIER , N. Introduction to the Bio-Entity Recognition Task at JNLPBA. In Proceedings of the International Joint Workshop on Natural Language Processing in Biomedicine and its Applications (2004), Citeseer, pp. 70–75

2004
[123]

FEVER breaker’s run of team NbAuzDrLqg

KIM, Y., AND ALLAN , J. FEVER breaker’s run of team NbAuzDrLqg. In Proceedings of the Second Workshop on Fact Extraction and VERification (FEVER) (Hong Kong, China, Nov. 2019), Association for Computational Linguistics, pp. 99–104

2019
[124]

Domain Attention With an Ensemble of Experts

KIM, Y.-B., S TRATOS , K., AND KIM, D. Domain Attention With an Ensemble of Experts. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (2017), pp. 643–653

2017
[125]

Investigating the Influence of Noise and Distractors on the Interpretation of Neural Networks

KINDERMANS , P.-J., S CHÜTT , K., M ÜLLER , K.-R., AND DÄHNE , S. Investigating the Influence of Noise and Distractors on the Interpretation of Neural Networks. arXiv preprint arXiv:1611.07270 (2016)

2016 arXiv
[126]

C., AND SUGIYAMA , M

KIRYO, R., N IU, G., DU PLESSIS , M. C., AND SUGIYAMA , M. Positive-Unlabeled Learning With Non-Negative Risk Estimator. In Advances in Neural Information Processing Systems (2017), pp. 1675–1685

2017
[127]

Towards Automated Factchecking: Developing an Annotation Schema and Benchmark For Consistent Automated Claim Detection

KONSTANTINOVSKIY , L., P RICE , O., B ABAKAR , M., AND ZUBIAGA , A. Towards Automated Factchecking: Developing an Annotation Schema and Benchmark For Consistent Automated Claim Detection. arXiv preprint arXiv:1809.08193 (2018)

2018 arXiv
[128]

Explainable Automated Fact-Checking for Public Health Claims

KOTONYA , N., AND TONI , F. Explainable Automated Fact-Checking for Public Health Claims. In Proceedings of the 2020 Conference on Empirical Methods in Natural Lan- guage Processing (EMNLP) (Online, Nov. 2020), Association for Computational Linguistics, pp. 7740–7754

2020
[129]

M., AND LOOG , M

KOUW, W. M., AND LOOG , M. A Review of Domain Adaptation Without Target Labels. IEEE Transactions on Pattern Analysis and Machine Intelligence (2019)

2019
[130]

K., B RUNAK , S., L UND, O., O PREA , T

KRINGELUM , J., K JAERULFF , S. K., B RUNAK , S., L UND, O., O PREA , T. I., AND TABOUREAU , O. ChemProt-3.0: a Global Chemical Biology Diseases Mapping. Database 2016 (2016). 144

2016
[131]

Computing Krippendorff’s alpha-reliability

K RIPPENDORFF , K. Computing Krippendorff’s alpha-reliability
[132]

KUA, E., R EDER , M., AND GROSSEL , M. J. Science in the news: a study of reporting genomics. Public Understanding of Science 13, 3 (2004), 309–322

2004
[133]

E., NAGY, M

KUHN , T., B ARBANO , P . E., NAGY, M. L., AND KRAUTHAMMER , M. Broadening the Scope of Nanopublications. In The Semantic Web: Semantics and Big Data, 10th International Conference, ESWC 2013, Montpellier, France, May 26-30, 2013. Proceedings (2013), P . Cimiano, Ó. Corcho, V....

2013
[134]

What Y ou Saw is Not What Y ou Get: Domain Adaptation Using Asymmetric Kernel Transforms

KULIS , B., S AENKO , K., AND DARRELL , T. What Y ou Saw is Not What Y ou Get: Domain Adaptation Using Asymmetric Kernel Transforms. In CVPR (2011), IEEE Computer Society, pp. 1785–1792

2011
[135]

S., W INNEG , K., HALL JAMIESON , K., AND ALBARRACÍN , D

KURU, O., S TECULA , D., L U, H., O PHIR , Y., C HAN , M.- P. S., W INNEG , K., HALL JAMIESON , K., AND ALBARRACÍN , D. The effects of scientific messages and narratives about vaccination. PLoS One 16, 3 (2021), e0248328

2021
[136]

LEHMAN , E., D EYOUNG , J., B ARZILAY, R., AND WALLACE , B. C. Inferring which medical treatments work from reports of clinical trials. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technolo...

2019
[137]

Learning From Positive and Unlabeled Examples

LETOUZEY , F., D ENIS , F., AND GILLERON , R. Learning From Positive and Unlabeled Examples. In International Conference on Algorithmic Learning Theory (2000), Springer, pp. 71–85

2000
[138]

BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension

LEWIS , M., L IU, Y., GOYAL, N., G HAZVININEJAD , M., M OHAMED , A., L EVY, O., S TOYANOV, V., AND ZETTLEMOYER , L. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension. In Proceedings of the 58th Annual Meeting of t...

2020
[139]

Spotting Fake Reviews Via Collective Positive-Unlabeled Learning

LI, H., C HEN , Z., L IU, B., W EI, X., AND SHAO, J. Spotting Fake Reviews Via Collective Positive-Unlabeled Learning. In 2014 IEEE International Conference on Data Mining (2014), IEEE, pp. 899–904

2014
[140]

J., S CIAKY, D., W EI, C.-H., L EAMAN , R., D AVIS, A

LI, J., S UN, Y., J OHNSON , R. J., S CIAKY, D., W EI, C.-H., L EAMAN , R., D AVIS, A. P., MATTINGLY, C. J., W IEGERS , T. C., AND LU, Z. BioCreative V CDR task corpus: a resource for chemical disease relation extraction. Database 2016 (2016). 145

2016
[141]

A., AND PENG , N

LI, X., B URNS , G. A., AND PENG , N. A Paragraph-level Multi-task Learning Model for Scientific Fact-Verification. In Proceedings of the Workshop on Scientific Document Understanding co-located with 35th AAAI Conference on Artificial Inteligence, SDU@AAAI 2021, Virtual Event,...

2021
[142]

What’s in a Domain? Learning Domain-Robust Text Representations Using Adversarial Training

LI, Y., BALDWIN , T., AND COHN , T. What’s in a Domain? Learning Domain-Robust Text Representations Using Adversarial Training. 474–479
[143]

An NLP Analysis of Exaggerated Claims in Science News

LI, Y., Z HANG , J., AND YU, B. An NLP Analysis of Exaggerated Claims in Science News. In Proceedings of the 2017 Workshop: Natural Language Processing meets Journalism, NLPmJ@EMNLP , Copenhagen, Denmark, September 7, 2017 (2017), O. Popescu and C. Strapparava, Eds., Associati...

2017
[144]

LIN, C., B ETHARD , S., D LIGACH , D., S ADEQUE , F., SAVOVA, G., AND MILLER , T. A. Does BERT Need Domain Adaptation for Clinical Negation Detection? Journal of the American Medical Informatics Association 27, 4 (2020), 584–591

2020
[145]

Rouge: A Package for Automatic Evaluation of Summaries

LIN, C.-Y. Rouge: A Package for Automatic Evaluation of Summaries. In Text summariza- tion branches out (2004), pp. 74–81

2004
[146]

C., W ANG , Y.-X., AND SMOLA , A

LIPTON , Z. C., W ANG , Y.-X., AND SMOLA , A. J. Detecting and Correcting for Label Shift with Black Box Predictors. In ICML (2018), J. G. Dy and A. Krause, Eds., vol. 80 of Proceedings of Machine Learning Research, PMLR, pp. 3128–3136

2018
[147]

Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing

LIU, P., YUAN, W., F U, J., J IANG , Z., H AYASHI, H., AND NEUBIG , G. Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing. CoRR abs/2107.13586 (2021)

2021 arXiv
[148]

RoBERTa: A Robustly Optimized BERT Pretraining Approach

LIU, Y., O TT, M., G OYAL, N., D U, J., J OSHI , M., C HEN , D., L EVY, O., L EWIS , M., ZETTLEMOYER , L., AND STOYANOV, V. RoBERTa: A Robustly Optimized BERT Pretraining Approach. CoRR abs/1907.11692 (2019)

2019 arXiv
[149]

L., N EUMANN , M., K INNEY, R., AND WELD, D

LO, K., W ANG , L. L., N EUMANN , M., K INNEY, R., AND WELD, D. S. S2ORC: The Semantic Scholar Open Research Corpus. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (2020), pp. 4969–4983

2020
[150]

Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction

LUAN, Y., HE, L., O STENDORF , M., AND HAJISHIRZI , H. Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (Brussels, Belgium, O...

2018
[151]

R., AND WONG , K

MA, J., G AO, W., J OTY, S. R., AND WONG , K. Sentence-level evidence embedding for claim verification with hierarchical attention networks. In Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 20...

2019
[152]

Detect Rumors on Twitter by Promoting Information Campaigns With Generative Adversarial Learning

MA, J., G AO, W., AND WONG , K.-F. Detect Rumors on Twitter by Promoting Information Campaigns With Generative Adversarial Learning. In The World Wide Web Conference (2019), pp. 3049–3055

2019
[153]

Domain Adaptation with BERT -based Domain Classification and Data Selection

MA, X., X U, P., W ANG , Z., N ALLAPATI , R., AND XIANG , B. Domain Adaptation with BERT -based Domain Classification and Data Selection. InProceedings of the 2nd Workshop on Deep Learning Approaches for Low-Resource NLP (DeepLo 2019) (2019), pp. 76–83

2019
[154]

MACKAY, D. J. C. Information-based objective functions for active data selection. Neural Comput. 4, 4 (1992), 590–604

1992
[155]

K., Z ETTLEMOYER , L., H ENDERSON , J., S AEIDI , M., M ATHIAS , L., S TOY- ANOV, V., AND YAZDANI , M

MAHABADI , R. K., Z ETTLEMOYER , L., H ENDERSON , J., S AEIDI , M., M ATHIAS , L., S TOY- ANOV, V., AND YAZDANI , M. PERFECT: prompt-free and efficient few-shot learning with language models. CoRR abs/2204.01172 (2022)

2022 arXiv
[156]

Team Papelo: Transformer Networks at FEVER

MALON , C. Team Papelo: Transformer Networks at FEVER. In Proceedings of the First Workshop on Fact Extraction and VERification (FEVER) (2018), pp. 109–113

2018
[157]

P ., SANTORINI , B., AND MARCINKIEWICZ , M

MARCUS , M. P ., SANTORINI , B., AND MARCINKIEWICZ , M. A. Building a Large Annotated Corpus of English: The Penn Treebank. Comput. Linguistics 19, 2 (1993), 313–330

1993
[158]

M., AND JARVIS , S

MCCARTHY, P. M., AND JARVIS , S. Mtld, vocd-d, and hd-d: A validation study of sophisticated approaches to lexical diversity assessment. Behavior research methods 42, 2 (2010), 381–392

2010
[159]

L., P ITOUT, I

MEJZINI , R., F LYNN, L. L., P ITOUT, I. L., F LETCHER , S., W ILTON, S. D., AND AKKARI , P. A. ALS Genetics, Mechanisms, and Therapeutics: Where Are We Now? Frontiers in Neuroscience 13 (2019)

2019
[160]

MENCHER , M., AND SHILTON , W. P. News reporting and writing. Brown & Benchmark Publishers Madison, WI, 1997

1997
[161]

MICHEL , P., L I, X., N EUBIG , G., AND PINO, J. M. On Evaluation of Adversarial Per- turbations for Sequence-to-Sequence Models. In Proceedings of NAACL-HLT (2019), pp. 3103–3114. 147

2019
[162]

MedMentions: A Large Biomedical Corpus Annotated with UMLS Concepts

MOHAN , S., AND LI, D. MedMentions: A Large Biomedical Corpus Annotated with UMLS Concepts. In 1st Conference on Automated Knowledge Base Construction, AKBC 2019, Amherst, MA, USA, May 20-22, 2019 (2019)

2019
[163]

Covert: A corpus of fact-checked biomedical COVID-19 tweets

MOHR , I., W ÜHRL , A., AND KLINGER , R. Covert: A corpus of fact-checked biomedical COVID-19 tweets. CoRR abs/2204.12164 (2022)

2022 arXiv
[164]

Bad science in the headlines: Who takes responsibility when science is distorted in the mass media? EMBO reports 7, 12 (2006), 1193–1196

MOORE , A. Bad science in the headlines: Who takes responsibility when science is distorted in the mass media? EMBO reports 7, 12 (2006), 1193–1196

2006
[165]

A., H AJISHIRZI , H., AND HOPE , T

NADKARNI , R., W ADDEN , D., B ELTAGY, I., S MITH , N. A., H AJISHIRZI , H., AND HOPE , T. Scientific language models for biomedical knowledge base completion: An empirical study. In 3rd Conference on Automated Knowledge Base Construction, AKBC 2021, Virtual, October 4-8, 2021...

2021
[166]

NAKOV, P ., MARTINO , G. D. S., E LSAYED , T., BARRÓN -CEDEÑO , A., M ÍGUEZ , R., S HAAR , S., A LAM , F., H AOUARI , F., H ASANAIN , M., B ABULKOV , N., N IKOLOV, A., S HAHI , G. K., STRUSS , J. M., AND MANDL , T. The CLEF-2021 CheckThat! Lab on Detecting Check- Worthy Claims...

2021
[167]

Communicating science effectively: A research agenda

NATIONAL ACADEMIES OF SCIENCES , ENGINEERING , AND MEDICINE . Communicating science effectively: A research agenda
[168]

Selling Science: How the Press Covers Science and Technology

N ELKIN , D. Selling Science: How the Press Covers Science and Technology
[169]

ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing

NEUMANN , M., K ING , D., B ELTAGY, I., AND AMMAR , W. ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing. In Proceedings of the 18th BioNLP Workshop and Shared Task, BioNLP@ACL 2019, Florence, Italy, August 1, 2019 (2019), D. Demner-Fushman, K. B. Coh...

2019
[170]

Adversarial NLI: A new benchmark for natural language understanding

NIE, Y., W ILLIAMS , A., D INAN , E., B ANSAL , M., W ESTON , J., AND KIELA , D. Adversarial NLI: A new benchmark for natural language understanding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (Online, 2020), Association for Comp...

2020
[171]

On a Generalization of the Jensen-Shannon Divergence and the Jensen- Shannon Centroid

NIELSEN , F. On a Generalization of the Jensen-Shannon Divergence and the Jensen- Shannon Centroid. Entropy 22, 2 (2020), 221. 148

2020
[172]

GEM: Generative enhanced model for adversarial attacks

NIEWINSKI , P., P SZONA , M., AND JANICKA , M. GEM: Generative enhanced model for adversarial attacks. In Proceedings of the Second Workshop on Fact Extraction and VERification (FEVER) (Hong Kong, China, Nov. 2019), Association for Computational Linguistics, pp. 20–26

2019
[173]

Improving paraphrase detection with the adversarial paraphrasing task

NIGHOJKAR , A., AND LICATO, J. Improving paraphrase detection with the adversarial paraphrasing task. In Proceedings of the 59th Annual Meeting of the Association for Com- putational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume...

2021
[174]

J., P ATEL, R., Y ANG , Y., M ARSHALL , I

N YE, B., L I, J. J., P ATEL, R., Y ANG , Y., M ARSHALL , I. J., N ENKOVA , A., AND WALLACE , B. C. A Corpus With Multi-Level Annotations of Patients, Interventions and Outcomes to Support Language Processing for Medical Literature. In Proceedings of the conference. Associatio...

2018
[175]

Multi-Hop Fact Checking of Political Claims

OSTROWSKI , W., A RORA , A., A TANASOVA , P., AND AUGENSTEIN , I. Multi-Hop Fact Checking of Political Claims. arXiv preprint arXiv:2009.06401 (2020)

2020 arXiv
[176]

PAN, L., C HEN , W., X IONG , W., K AN, M., AND WANG , W. Y. Zero-shot Fact Verification by Claim Generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, ...

2021
[177]

Comparing Bayesian Models of Annotation

PAUN, S., C ARPENTER , B., C HAMBERLAIN , J., H OVY, D., K RUSCHWITZ , U., AND POESIO , M. Comparing Bayesian Models of Annotation. Trans. Assoc. Comput. Linguistics 6 (2018), 571–585

2018
[178]

Potato: The portable text annotation tool

PEI, J., A NANTHASUBRAMANIAM , A., W ANG , X., Z HOU, N., D EDELOUDIS , A., S ARGENT , J., AND JURGENS , D. Potato: The portable text annotation tool. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations (2022)

2022
[179]

Measuring sentence-level and aspect-level (un)certainty in science communications

PEI, J., AND JURGENS , D. Measuring sentence-level and aspect-level (un)certainty in science communications. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 7-11 November, 20...

2021
[180]

PELLECHIA , M. G. Trends in science coverage: A content analysis of three us newspapers. Public Understanding of Science 6, 1 (1997), 49

1997
[181]

Distantly Supervised Named Entity Recognition using Positive-Unlabeled Learning

PENG , M., X ING , X., Z HANG , Q., F U, J., AND HUANG , X.-J. Distantly Supervised Named Entity Recognition using Positive-Unlabeled Learning. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (2019), pp. 2409–2419

2019
[182]

E., N EUMANN , M., I YYER , M., G ARDNER , M., C LARK , C., L EE, K., AND ZETTLEMOYER , L

PETERS , M. E., N EUMANN , M., I YYER , M., G ARDNER , M., C LARK , C., L EE, K., AND ZETTLEMOYER , L. Deep contextualized word representations. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language...

2018
[183]

C., B ATTLEDAY, R

PETERSON , J. C., B ATTLEDAY, R. M., G RIFFITHS , T. L., AND RUSSAKOVSKY , O. Human Uncertainty Makes Classification More Robust. In 2019 IEEE/CVF International Conference on Computer Vision, ICCV 2019, Seoul, Korea (South), October 27 - November 2, 2019 (2019), IEEE, pp. 9616–9625

2019
[184]

PLANK , B., H OVY, D., AND SØGAARD , A. Linguistically debatable or just plain wrong? In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014, June 22-27, 2014, Baltimore, MD, USA, Volume 2: Short Papers (2014), The Association for ...

2014
[185]

POLIAK , A., N ARADOWSKY , J., H ALDAR , A., R UDINGER , R., AND DURME , B. V. Hy- pothesis Only Baselines in Natural Language Inference. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, *SEM@NAACL-HLT 2018, New Orleans, Louisiana, USA, Ju...

2018
[186]

P., H AGEN , M., AND STEIN , B

POTTHAST , M., G OLLUB , T., K OMLOSSY , K., S CHUSTER , S., W IEGMANN , M., GARCES FERNANDEZ , E. P., H AGEN , M., AND STEIN , B. Crowdsourcing a large corpus of clickbait on Twitter. In Proceedings of the 27th International Conference on Computa- tional Linguistics (Santa Fe...

2018
[187]

News and its Communicative Quality: The Inverted Pyramid—When and Why Did it Appear? Journalism Studies 4, 4 (2003), 501–511

PÖTTKER , H. News and its Communicative Quality: The Inverted Pyramid—When and Why Did it Appear? Journalism Studies 4, 4 (2003), 501–511

2003
[188]

L., M CFARLAND , D., AND JURAFSKY , D

PRABHAKARAN , V., H AMILTON , W. L., M CFARLAND , D., AND JURAFSKY , D. Predicting the rise and fall of scientific topics from trends in their rhetorical framing. In Proceedings of 150 the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Pap...

2016
[189]

Scientific Claim Verification with VERT5ERINI

PRADEEP, R., M A, X., N OGUEIRA , R., AND LIN, J. Scientific Claim Verification with VERT5ERINI. CoRR abs/2010.11930 (2020)

2020 arXiv
[190]

Zero-shot Text Classification With Generative Language Models

PURI , R., AND CATANZARO , B. Zero-shot Text Classification With Generative Language Models. CoRR abs/1912.10165 (2019)

2019 arXiv
[191]

Improving Language Understanding by Generative Pre-Training

RADFORD , A., N ARASIMHAN , K., S ALIMANS , T., AND SUTSKEVER , I. Improving Language Understanding by Generative Pre-Training. Technical report
[192]

Language models are unsupervised multitask learners

RADFORD , A., W U, J., C HILD, R., L UAN, D., A MODEI , D., AND SUTSKEVER , I. Language models are unsupervised multitask learners. OpenAI Blog 1, 8 (2019), 9

2019
[193]

RAFFEL , C., S HAZEER , N., R OBERTS , A., L EE, K., N ARANG , S., M ATENA , M., Z HOU, Y., LI, W., AND LIU, P. J. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer. J. Mach. Learn. Res. 21 (2020), 140:1–140:67

2020
[194]

SQuAD: 100, 000+ Questions for Machine Comprehension of Text

RAJPURKAR , P., ZHANG , J., L OPYREV , K., AND LIANG , P. SQuAD: 100, 000+ Questions for Machine Comprehension of Text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, EMNLP 2016, Austin, Texas, USA, November 1-4, 2016 (2016), J. Su, ...

2016
[195]

Dietary magnesium tied to lower risk of heart disease and diabetes

RAPPAPORT , L. Dietary magnesium tied to lower risk of heart disease and diabetes. Reuters
[196]

Citation Needed: A Taxonomy and Algorithmic Assessment of Wikipedia’s Verifiability

REDI , M., F ETAHU, B., M ORGAN , J., AND TARABORELLI , D. Citation Needed: A Taxonomy and Algorithmic Assessment of Wikipedia’s Verifiability. In The World Wide Web Conference (2019), pp. 1567–1578

2019
[197]

Sentence-BERT: Sentence embeddings using Siamese BERT-networks

REIMERS , N., AND GUREVYCH , I. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) (H...

2019
[198]

Generating natural language adversarial examples through probability weighted word saliency

REN, S., D ENG , Y., H E, K., AND CHE, W. Generating natural language adversarial examples through probability weighted word saliency. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (2019), pp. 1085–1097. 151

2019
[199]

Positive Unlabeled Learning for Deceptive Reviews Detection

REN, Y., J I, D., AND ZHANG , H. Positive Unlabeled Learning for Deceptive Reviews Detection. In EMNLP 2014 (2014), pp. 488–498

2014
[200]

Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering

RIABI , A., S CIALOM , T., K ERARON , R., S AGOT, B., S EDDAH , D., AND STAIANO , J. Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering. In Pro- ceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Eve...

2021
[201]

T., S INGH , S., AND GUESTRIN , C

RIBEIRO , M. T., S INGH , S., AND GUESTRIN , C. Semantically equivalent adversarial rules for debugging nlp models. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (2018), pp. 856–865

2018
[202]

Writing strategies: Reaching diverse audiences, vol

RICHARDSON , L. Writing strategies: Reaching diverse audiences, vol. 21. Sage Publica- tions, 1990

1990
[203]

Adapt or Get Left Behind: Domain Adaptation Through Bert Language Model Finetuning for Aspect-Target Sentiment Classification

RIETZLER , A., S TABINGER , S., O PITZ , P., AND ENGL , S. Adapt or Get Left Behind: Domain Adaptation Through Bert Language Model Finetuning for Aspect-Target Sentiment Classification. 4933–4941
[204]

E., W ALKER , S., J ONES , S., H ANCOCK -BEAULIEU , M., AND GATFORD, M

ROBERTSON , S. E., W ALKER , S., J ONES , S., H ANCOCK -BEAULIEU , M., AND GATFORD, M. Okapi at trec-3. In TREC (1994)

1994
[205]

Quality and integrity in scientific writing: prerequisites for quality in science communication

ROLAND , M.-C. Quality and integrity in scientific writing: prerequisites for quality in science communication. Journal of Science Communication 8, 2 (2009), A04

2009
[206]

COVID-Fact: Fact Extraction and Verification of Real-World Claims on COVID-19 Pandemic

SAAKYAN , A., C HAKRABARTY , T., AND MURESAN , S. COVID-Fact: Fact Extraction and Verification of Real-World Claims on COVID-19 Pandemic. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on N...

2021
[207]

NegatER: Unsupervised Discovery of Negatives in Commonsense Knowledge Bases

SAFAVI, T., Z HU, J., AND KOUTRA , D. NegatER: Unsupervised Discovery of Negatives in Commonsense Knowledge Bases. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Do- minican Republic, 7-11 Nove...

2021
[208]

J., S OMMERFELDT , E

SAFFER , A. J., S OMMERFELDT , E. J., AND TAYLOR , M. The effects of organizational twitter interactivity on organization–public relationships. Public relations review 39, 3 (2013), 213–215. 152

2013
[209]

B., M OHANKUMAR , A

SAI, A. B., M OHANKUMAR , A. K., AND KHAPRA , M. M. A Survey of Evaluation Metrics Used for NLG Systems. CoRR abs/2008.12009 (2020)

2020 arXiv
[210]

SALITA , J. T. Writing for lay audiences: A challenge for scientists. Medical Writing 24 (2015), 183–189

2015
[211]

DistilBERT, a Distilled Version of BERT: Smaller, Faster, Cheaper and Lighter

SANH , V., D EBUT, L., C HAUMOND , J., AND WOLF, T. DistilBERT, a Distilled Version of BERT: Smaller, Faster, Cheaper and Lighter. arXiv preprint arXiv:1910.01108 (2019)

2019 arXiv
[212]

Automatically Identifying Words That Can Serve as Labels for Few-Shot Text Classification

SCHICK , T., S CHMID , H., AND SCHÜTZE , H. Automatically Identifying Words That Can Serve as Labels for Few-Shot Text Classification. InProceedings of the 28th International Conference on Computational Linguistics, COLING 2020, Barcelona, Spain (Online), De- cember 8-13, 2020...

2020
[213]

Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference

SCHICK , T., AND SCHÜTZE , H. Exploiting Cloze Questions for Few-Shot Text Classification and Natural Language Inference. Computing Research Repository arXiv:2001.07676 (2020)

2020 arXiv
[214]

It’s Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners

SCHICK , T., AND SCHÜTZE , H. It’s Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners. Computing Research Repository arXiv:2009.07118 (2020)

2020 arXiv
[215]

S., K ARPUKHIN , V., O GUZ , B., L EWIS , M., Y IH, W., AND RIEDEL , S

SCHLICHTKRULL , M. S., K ARPUKHIN , V., O GUZ , B., L EWIS , M., Y IH, W., AND RIEDEL , S. Joint verification and reranking for open fact checking over tables. InProceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International ...

2021
[216]

That is a known lie: Detecting previously fact-checked claims

SHAAR , S., B ABULKOV , N., D A SAN MARTINO , G., AND NAKOV, P. That is a known lie: Detecting previously fact-checked claims. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (Online, July 2020), Association for Computational Linguist...

2020
[217]

SHIN , T., R AZEGHI , Y., IV, R. L. L., W ALLACE , E., AND SINGH , S. Autoprompt: Eliciting knowledge from language models with automatically generated prompts. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Online, Novem...

2020
[218]

P., N ELSON , L

SIMMONS , J. P., N ELSON , L. D., AND SIMONSOHN , U. False-positive psychology: undis- closed flexibility in data collection and analysis allows presenting anything as significant. 153
[219]

SNOW, R., O’C ONNOR , B., J URAFSKY , D., AND NG, A. Y. Cheap and Fast - But is it Good? Evaluating Non-Expert Annotations for Natural Language Tasks. In 2008 Conference on Empirical Methods in Natural Language Processing, EMNLP 2008, Proceedings of the Conference, 25-27 Octob...

2008
[220]

Mpnet: Masked and permuted pre- training for language understanding

SONG , K., T AN, X., Q IN, T., L U, J., AND LIU, T. Mpnet: Masked and permuted pre- training for language understanding. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020,...

2020
[221]

Supervised Keyphrase Extraction as Positive Unlabeled Learning

STERCKX , L., D EMEESTER , T., DEVELDER , C., AND CARAGEA , C. Supervised Keyphrase Extraction as Positive Unlabeled Learning. In EMNLP 2016 (2016), pp. 1–6

2016
[222]

K., S AMARAS , C., O’N EILL , E

STOLAROFF , J. K., S AMARAS , C., O’N EILL , E. R., L UBERS , A., M ITCHELL , A. S., AND CEPERLEY , D. Energy use and life cycle greenhouse gas emissions of drones for commercial package delivery. Nature communications 9, 1 (2018), 1–13

2018
[223]

SUGIYAMA , K., K UMAR , T., KAN, M.-Y., AND TRIPATHI , R. C. Identifying Citing Sentences in Research Papers Using Supervised Learning. In 2010 International Conference on Information Retrieval & Knowledge Management (CAMP) (2010), IEEE, pp. 67–72

2010
[224]

A., D AVIES , A., O GDEN , J., W HELAN , L., H UGHES , B., D ALTON, B., ET AL

SUMNER , P., VIVIAN -GRIFFITHS , S., B OIVIN , J., W ILLIAMS , A., V ENETIS , C. A., D AVIES , A., O GDEN , J., W HELAN , L., H UGHES , B., D ALTON, B., ET AL . The Association Between Exaggeration in Health Related Science News and Academic Press Releases: Retrospective Obser...

2014
[225]

Return of Frustratingly Easy Domain Adaptation

SUN, B., F ENG , J., AND SAENKO , K. Return of Frustratingly Easy Domain Adaptation. In AAAI (2016), D. Schuurmans and M. P . Wellman, Eds., AAAI Press, pp. 2058–2065

2016
[226]

Intriguing properties of neural networks

SZEGEDY, C., Z AREMBA , W., S UTSKEVER , I., B RUNA , J., E RHAN , D., G OODFELLOW , I., AND FERGUS , R. Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199 (2013)

2013 arXiv
[227]

A corpus of sentence-level revisions in academic writing: A step towards understanding statement strength in communication

TAN, C., AND LEE, L. A corpus of sentence-level revisions in academic writing: A step towards understanding statement strength in communication. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (Baltimore, Mary...

2014
[228]

W., L ONG , M., A SHLEY, E., D ENNING , A., G OUT, B., H ANSEN , K., H UWS, T., JENNINGS , L., Q UINN , S., S ARKIES , P., ET AL

TAYLOR , J. W., L ONG , M., A SHLEY, E., D ENNING , A., G OUT, B., H ANSEN , K., H UWS, T., JENNINGS , L., Q UINN , S., S ARKIES , P., ET AL . When medical news comes from press 154 releases—a case study of pancreatic cancer and processed meat. PloS one 10, 6 (2015), e0127848

2015
[229]

Automated Fact Checking: Task Formulations, Methods and Future Directions

THORNE , J., AND VLACHOS , A. Automated Fact Checking: Task Formulations, Methods and Future Directions. In Proceedings of the 27th International Conference on Computa- tional Linguistics (Santa Fe, New Mexico, USA, Aug. 2018), Association for Computational Linguistics, pp. 3346–3359

2018
[230]

FEVER: a Large-scale Dataset for Fact Extraction and VERification

THORNE , J., V LACHOS , A., C HRISTODOULOPOULOS , C., AND MITTAL , A. FEVER: a Large-scale Dataset for Fact Extraction and VERification. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technol...

2018
[231]

Evaluating adversarial attacks against multiple fact verification systems

THORNE , J., V LACHOS , A., C HRISTODOULOPOULOS , C., AND MITTAL , A. Evaluating adversarial attacks against multiple fact verification systems. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference...

2019
[232]

The FEVER2.0 shared task

THORNE , J., V LACHOS , A., C OCARASCU , O., C HRISTODOULOPOULOS , C., AND MITTAL , A. The FEVER2.0 shared task. In Proceedings of the Second Workshop on Fact Extraction and VERification (FEVER) (Hong Kong, China, Nov. 2019), Association for Computational Linguistics, pp. 1–6

2019
[233]

Scaling and Disagreements: Bias, Noise, and Ambiguity

UMA, A., A LMANEA , D., AND POESIO , M. Scaling and Disagreements: Bias, Noise, and Ambiguity. Frontiers Artif. Intell. 5 (2022), 818451

2022
[234]

A Case for Soft Loss Functions

UMA, A., F ORNACIARI , T., H OVY, D., PAUN, S., P LANK , B., AND POESIO , M. A Case for Soft Loss Functions. In Proceedings of the Eighth AAAI Conference on Human Computation and Crowdsourcing, HCOMP 2020, Hilversum, The Netherlands (virtual), October 25-29, 2020 (2020), L. Ar...

2020
[235]

Learning from Disagreement: A Survey

UMA, A., F ORNACIARI , T., H OVY, D., PAUN, S., P LANK , B., AND POESIO , M. Learning from Disagreement: A Survey. J. Artif. Intell. Res. 72 (2021), 1385–1470

2021
[236]

V., AND VARMA , V

VADAPALLI , R., S YED, B., P RABHU , N., S RINIVASAN , B. V., AND VARMA , V. When science journalism meets artificial intelligence : An interactive demonstration. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demon- strations...

2018
[237]

N., KAISER , Ł., AND POLOSUKHIN , I

VASWANI , A., S HAZEER , N., P ARMAR , N., U SZKOREIT , J., J ONES , L., G OMEZ , A. N., KAISER , Ł., AND POLOSUKHIN , I. Attention is All Y ou Need. In Advances in Neural Information Processing Systems (2017), pp. 5998–6008

2017
[238]

Fact Checking: Task Definition and Dataset Construction

VLACHOS , A., AND RIEDEL , S. Fact Checking: Task Definition and Dataset Construction. In Proceedings of the ACL 2014 Workshop on Language Technologies and Computational Social Science (2014), pp. 18–22

2014
[239]

L., VAN ZUYLEN , M., C OHAN , A., AND HA- JISHIRZI , H

WADDEN , D., L IN, S., L O, K., W ANG , L. L., VAN ZUYLEN , M., C OHAN , A., AND HA- JISHIRZI , H. Fact or Fiction: Verifying Scientific Claims. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Online, November 16-20, 2020 ...

2020
[240]

L., C OHAN , A., B ELTAGY, I., AND HAJISHIRZI , H

WADDEN , D., L O, K., W ANG , L. L., C OHAN , A., B ELTAGY, I., AND HAJISHIRZI , H. Longchecker: Improving scientific claim verification by modeling full-abstract context. CoRR abs/2112.01640 (2021)

2021 arXiv
[241]

Universal Adversarial Triggers for Attacking and Analyzing NLP

WALLACE , E., F ENG , S., K ANDPAL , N., G ARDNER , M., AND SINGH , S. Universal Adversarial Triggers for Attacking and Analyzing NLP. InProceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural...

2019
[242]

One step forward, one step back: changes in news coverage of medical interventions

WALSH -CHILDERS , K., B RADDOCK , J., R ABAZA , C., AND SCHWITZER , G. One step forward, one step back: changes in news coverage of medical interventions. Health commu- nication 33, 2 (2018), 174–187

2018
[243]

Medical Relation Extraction with Manifold Models

WANG , C., AND FAN, J. Medical Relation Extraction with Manifold Models. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014, June 22-27, 2014, Baltimore, MD, USA, Volume 1: Long Papers (2014), The Association for Computer Lingu...

2014
[244]

Minilm: Deep self- attention distillation for task-agnostic compression of pre-trained transformers

WANG , W., W EI, F., D ONG , L., B AO, H., Y ANG , N., AND ZHOU, M. Minilm: Deep self- attention distillation for task-agnostic compression of pre-trained transformers. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Pro- cessin...

2020
[245]

Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers, 2020

WANG , W., W EI, F., D ONG , L., B AO, H., Y ANG , N., AND ZHOU, M. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers, 2020. 156

2020
[246]

Liar, Liar Pants on Fire

WANG , W. Y. "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, ACL 2017, Vancouver, Canada, July 30 - August 4, Volume 2: Short Papers (2017), R. Barzilay and...

2017
[247]

A., A DELANI , D

WANG , Z., H ALE , S. A., A DELANI , D. I., G RABOWICZ , P. A., H ARTMANN , T., F LÖCK , F., AND JURGENS , D. Demographic inference and representative population estimates from multilingual social media data. In The World Wide Web Conference, WWW 2019, San Francisco, CA, USA, ...

2019
[248]

P ., MATTINGLY, C

WEI, C., P ENG , Y., L EAMAN , R., D AVIS, A. P ., MATTINGLY, C. J., L I, J., W IEGERS , T. C., AND LU, Z. Assessing the state of the art in biomedical relation extraction: overview of the biocreative V chemical-disease relation (CDR) task. Database J. Biol. Databases Curation...

2016
[249]

WEIGOLD , M. F. Communicating science: A review of the literature. Science communica- tion 23, 2 (2001), 164–193

2001
[250]

WILLIAMS , A., N ANGIA , N., AND BOWMAN , S. R. A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NA...

2018
[251]

L., G UGGER , S., D RAME , M., L HOEST, Q., AND RUSH , A

WOLF, T., D EBUT, L., S ANH , V., C HAUMOND , J., D ELANGUE , C., M OI, A., C ISTAC, P., RAULT, T., L OUF, R., F UNTOWICZ , M., D AVISON , J., S HLEIFER , S., VON PLATEN , P., MA, C., J ERNITE , Y., P LU, J., X U, C., S CAO, T. L., G UGGER , S., D RAME , M., L HOEST, Q., AND R...

2020
[252]

WOLOSHIN , S., AND SCHWARTZ , L. M. Press Releases: Translating Research Into News. Jama 287, 21 (2002), 2856–2858

2002
[253]

M., C ASELLA , S

WOLOSHIN , S., S CHWARTZ , L. M., C ASELLA , S. L., K ENNEDY, A. T., AND LARSON , R. J. Press Releases by Academic Medical Centers: Not So Academic? Annals of Internal Medicine 150, 9 (2009), 613–618. 157

2009
[254]

Claim Check-Worthiness Detection as Positive Unla- belled Learning

WRIGHT , D., AND AUGENSTEIN , I. Claim Check-Worthiness Detection as Positive Unla- belled Learning. In Findings of the Association for Computational Linguistics: EMNLP 2020 (Online, Nov. 2020), Association for Computational Linguistics, pp. 476–488

2020
[255]

Transformer Based Multi-Source Domain Adaptation

WRIGHT , D., AND AUGENSTEIN , I. Transformer Based Multi-Source Domain Adaptation. In Proceedings of EMNLP (2020), Association for Computational Linguistics

2020
[256]

CiteWorth: Cite-Worthiness Detection for Improved Scientific Document Understanding

WRIGHT , D., AND AUGENSTEIN , I. CiteWorth: Cite-Worthiness Detection for Improved Scientific Document Understanding. In Findings of the Association for Computational Linguistics: ACL/IJCNLP 2021, Online Event, August 1-6, 2021 (2021), C. Zong, F . Xia, W. Li, and R. Navigli, ...

2021
[257]

Semi-supervised exaggeration detection of health science press releases

WRIGHT , D., AND AUGENSTEIN , I. Semi-supervised exaggeration detection of health science press releases. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (2021), pp. 10824–10836

2021
[258]

L., F AUX, N., AND BECK , K

WRIGHT, D., G ENTILE , A. L., F AUX, N., AND BECK , K. L. Bioact: Biomedical knowledge base construction using active learning. bioRxiv (2022)

2022
[259]

NormCo: Deep Disease Normal- ization for Biomedical Knowledge Base Construction

WRIGHT , D., K ATSIS, Y., MEHTA, R., AND HSU, C.-N. NormCo: Deep Disease Normal- ization for Biomedical Knowledge Base Construction. In AKBC (2019)

2019
[260]

WRIGHT, D., W ADDEN , D., L O, K., K UEHL , B., C OHAN , A., A UGENSTEIN , I., AND WANG , L. L. Generating scientific claims for zero-shot scientific fact checking. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Paper...

2022
[261]

Semi-supervised Domain Adaptation with Subspace Learning for Visual Recognition

YAO, T., PAN, Y., N GO, C.-W., L I, H., AND MEI, T. Semi-supervised Domain Adaptation with Subspace Learning for Visual Recognition. In CVPR (2015), IEEE Computer Society, pp. 2142–2150

2015
[262]

R., L I, I., F RIEDMAN , D., AND RADEV, D

YASUNAGA , M., K ASAI , J., Z HANG , R., F ABBRI , A. R., L I, I., F RIEDMAN , D., AND RADEV, D. R. Scisummnet: A large annotated corpus and content-impact models for scientific paper summarization with citation networks. In The Thirty-Third AAAI Conference on Artificial Intel...

2019
[263]

Misrepresentation of randomized controlled trials in press releases and news coverage: a cohort study

YAVCHITZ , A., B OUTRON , I., B AFETA , A., M ARROUN , I., C HARLES , P., MANTZ , J., AND RAVAUD, P. Misrepresentation of randomized controlled trials in press releases and news coverage: a cohort study
[264]

TwoWingOS: A Two-Wing Optimization Strategy for Evidential Claim Verification

YIN, W., AND ROTH, D. TwoWingOS: A Two-Wing Optimization Strategy for Evidential Claim Verification. In EMNLP (2018), E. Riloff, D. Chiang, J. Hockenmaier, and J. Tsujii, Eds., Association for Computational Linguistics, pp. 105–114

2018
[265]

Detecting Causal Language Use in Science Findings

YU, B., L I, Y., AND WANG , J. Detecting Causal Language Use in Science Findings. In EMNLP (2019), pp. 4656–4666

2019
[266]

Measuring Correlation-to-Causation Exaggeration in Press Releases

YU, B., W ANG , J., G UO, L., AND LI, Y. Measuring Correlation-to-Causation Exaggeration in Press Releases. In Proceedings of the 28th International Conference on Computational Linguistics (2020), pp. 4860–4872

2020
[267]

L., AND RANGARAJAN , A

YUILLE , A. L., AND RANGARAJAN , A. The Concave-Convex Procedure (CCCP). In Advances in Neural Information Processing Systems 14 [Neural Information Processing Systems: Natural and Synthetic, NIPS 2001, December 3-8, 2001, Vancouver, British Columbia, Canada] (2001), T. G. Die...

2001
[268]

Diffusion of scientific articles across online platforms

ZAKHLEBIN , I., AND HORVÁT, E.-A. Diffusion of scientific articles across online platforms. In Proceedings of the International AAAI Conference on Web and Social Media (2020), vol. 14, pp. 762–773

2020
[269]

E., S HENG , Q

ZHANG , W. E., S HENG , Q. Z., A LHAZMI , A., AND LI, C. Adversarial Attacks on Deep Learning Models in Natural Language Processing: A Survey, 2019

2019
[270]

Paraphrase Generation: A Survey of the State of the Art

ZHOU, J., AND BHAT, S. Paraphrase Generation: A Survey of the State of the Art. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021, Virtual Event / Punta Cana, Dominican Republic, 7-11 November, 2021 (2021), M. Moens, X. Huang...

2021
[271]

A comprehensive survey on transfer learning

ZHUANG , F., Q I, Z., D UAN, K., X I, D., Z HU, Y., Z HU, H., X IONG , H., AND HE, Q. A comprehensive survey on transfer learning. Proc. IEEE 109, 1 (2021), 43–76

2021
[272]

Neural Structural Correspondence Learning for Domain Adaptation

ZISER , Y., AND REICHART , R. Neural Structural Correspondence Learning for Domain Adaptation. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017) (2017), pp. 400–410. 159

2017
[273]

Pivot based language modeling for improved neural domain adaptation

ZISER , Y., AND REICHART , R. Pivot based language modeling for improved neural domain adaptation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2018, New Orleans, Lo...

2018
[274]

Detection and Resolution of Rumours in Social Media: A Survey

ZUBIAGA , A., A KER , A., B ONTCHEVA , K., L IAKATA, M., AND PROCTER , R. Detection and Resolution of Rumours in Social Media: A Survey. ACM Computing Surveys (CSUR) 51, 2 (2018), 1–36

2018
[275]

Discourse-Aware Rumour Stance Classification in Social Media Using Sequential Classifiers

ZUBIAGA , A., K OCHKINA , E., L IAKATA, M., P ROCTER , R., L UKASIK , M., B ONTCHEVA , K., COHN , T., AND AUGENSTEIN , I. Discourse-Aware Rumour Stance Classification in Social Media Using Sequential Classifiers. Information Processing & Management 54, 2 (2018), 273–290

2018
[276]

Exploiting Context for Rumour Detection in Social Media

ZUBIAGA , A., L IAKATA, M., AND PROCTER , R. Exploiting Context for Rumour Detection in Social Media. In International Conference on Social Informatics (2017), Springer, pp. 109– 123

2017
[277]

Hands up! Don’t shoot!

ZUBIAGA , A., L IAKATA, M., P ROCTER , R., H OI, G. W. S., AND TOLMIE , P. Analysing How People Orient to and Spread Rumours in Social Media by Looking at Conversational Threads. PloS one 11, 3 (2016). 160 A Appendices for Claim Check-Worthiness Detection as Pos- itive Unlabel...

2016
[278]

Split the test set in half, one half for validation and one half for test
[279]

Optimize a temperature parameter T to minimize the average negative log-likelihood − 1 n P i log ˜p(yi = y∗ i |xi), where ˜pi = softmax( li T ) and li is the logits of the classifier, on the validation half of the test set
[280]

being intrigued

Measure the temperature scaled log-likelihood on the test half of the test set. Following the suggestion from [7], we run this procedure 5 times on different splits of the test set and take the average test-half log-likelihood as the result. D.2 Visualization Here we plot the ...

2000
[281]

185 – Overconsumption of sugar leads to diabetes

The information in the findings is completely different • Sentences in this category have findings which say completely different information • The sentences may be on totally different topics – Overconsumption of sugar causes diabetes – Regular exercise improves heart health ...
[282]

The information in the findings is mostly different • The findings may talk about the same topic, but the actual information is mostly different; for example, these sentences convey mostly different information even though they talk about the same topic: – Overconsumption of s...
[283]

The information in the findings is somewhat similar • The findings are discussing relevant research outputs but there are some differences in the information conveyed. Here the difference is that (i) talks about the relationship between overconsumption of sugar and diabetes an...
[284]

the level of generality

The information in the findings is mostly the same • In this case there may be some changes in e.g. the level of generality. Additionally, one sentence may go into more detail than the other and add additional context, but the information is largely the same • Here the two fin...
[285]

X is associated with Y

The information in the findings is completely the same • In this case there is complete overlap in the information in the findings conveyed by the two sentences – Overconsumption of sugar leads to diabetes. – The researchers found that overconsumption of sugar leads to diabete...
[286]

balancing

curate data and build models for performing the exaggeration detection task in two different settings: as predicting the individual claim strengths and comparing, and as an inference task where a model is fed both findings and asked to predict if the reference finding is being...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

A Graph-Theoretic Embedding- Based Approach for Rumor Detection in Twitter

ABULAISH , M., K UMARI , N., F AZIL , M., AND SINGH , B. A Graph-Theoretic Embedding- Based Approach for Rumor Detection in Twitter. In IEEE/WIC/ACM International Conference on Web Intelligence (2019), pp. 466–470

2019

[2] [2]

Unsupervised Domain Clusters in Pretrained Language Models

AHARONI , R., AND GOLDBERG , Y. Unsupervised Domain Clusters in Pretrained Language Models. In ACL (2020)

2020

[3] [3]

Time-Aware Evidence Ranking for Fact- Checking

ALLEIN , L., A UGENSTEIN , I., AND MOENS , M.-F. Time-Aware Evidence Ranking for Fact- Checking. arXiv preprint arXiv:2009.06402 (2020)

arXiv 2020

[4] [4]

Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning

ALLEN -ZHU, Z., AND LI, Y. Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning. CoRR abs/2012.09816 (2020)

arXiv 2020

[5] [5]

S., T HORNE , J., V LACHOS , A., C HRISTODOULOPOU - LOS, C., C OCARASCU , O., AND MITTAL , A

ALY, R., G UO, Z., S CHLICHTKRULL , M. S., T HORNE , J., V LACHOS , A., C HRISTODOULOPOU - LOS, C., C OCARASCU , O., AND MITTAL , A. FEVEROUS: fact extraction and verification over unstructured and structured information. In Proceedings of the Neural Information Pro- cessing Systems Track on Datasets and Benchmarks 1, NeurIPS Datasets and Benchmarks 2021,...

2021

[6] [6]

Construction of the Literature Graph in Semantic Scholar

AMMAR , W., G ROENEVELD , D., B HAGAVATULA , C., B ELTAGY, I., C RAWFORD , M., D OWNEY, D., D UNKELBERGER , J., E LGOHARY , A., F ELDMAN , S., H A, V., ET AL . Construction of the Literature Graph in Semantic Scholar. NAACL HLT 2018 (2018), 84–91

2018

[7] [7]

ASHUKHA , A., L YZHOV, A., M OLCHANOV , D., AND VETROV, D. P. Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020 (2020), OpenReview.net

2020

[8] [8]

ATANASOVA , P., B ARRON -CEDENO , A., E LSAYED , T., S UWAILEH , R., Z AGHOUANI , W., KYUCHUKOV , S., M ARTINO , G. D. S., AND NAKOV, P .Overview of the CLEF-2018 CheckThat! Lab on Automatic Identification and Verification of Political Claims. Task 1: Check-Worthiness. arXiv preprint arXiv:1808.05542 (2018)

arXiv 2018