Filters








3,102 Hits in 5.7 sec

Automated Paraphrase Quality Assessment Using Language Models and Transfer Learning

Bogdan Nicula, Mihai Dascalu, Natalie N. Newton, Ellen Orcutt, Danielle S. McNamara
<span title="2021-12-06">2021</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/mvyhkzd6pbcqznrpy72k57bd6y" style="color: black;">Computers</a> </i> &nbsp;
Our study introduces and evaluates various machine learning models using handcrafted features combined with Extra Trees, Siamese neural networks using BiLSTM RNNs, and pretrained BERT-based models, together  ...  with transfer learning from a larger general paraphrase corpus, to estimate the quality of paraphrases across the four dimensions.  ...  Last, we trained the models on a generic paraphrase identification dataset (i.e., MSRP) and adapted them to the paraphrase quality assessment via transfer learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/computers10120166">doi:10.3390/computers10120166</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/krigfnhz55adnikkuum33nbppq">fatcat:krigfnhz55adnikkuum33nbppq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211208112006/https://mdpi-res.com/d_attachment/computers/computers-10-00166/article_deploy/computers-10-00166.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/de/87/de87f16ff997770b4b60843446eb496a7cea9f04.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/computers10120166"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

AGenT Zero: Zero-shot Automatic Multiple-Choice Question Generation for Skill Assessments [article]

Eric Li, Jingyi Su, Hao Sheng, Lawrence Wai
<span title="2020-12-18">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The automated generation of MCQs would allow assessment creation at scale. Recent advances in natural language processing have given rise to many complex question generation methods.  ...  In addition to maintaining semantic similarity between the question-answer pairs, our pipeline, which we call AGenT Zero, consists of only pre-trained models and requires no fine-tuning, minimizing data  ...  In this paper, Text-To-Text Transfer Transformer (T5) (Raffel et al. 2019 ) is used as our primary backbone model for question generation.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01186v2">arXiv:2012.01186v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/j7ru3aw7d5hqnez5573wy6v7xi">fatcat:j7ru3aw7d5hqnez5573wy6v7xi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201225055535/https://arxiv.org/pdf/2012.01186v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/8d/0b/8d0b2a373dc805d53a2e2c98b0c809bb29608c24.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.01186v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Development of Technology for Summarization of Kazakh Text

Talgat Zhabayev, Ualsher Tukeyev
<span title="">2021</span> <i title="The Science and Information Organization"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2yzw5hsmlfa6bkafwsibbudu64" style="color: black;">International Journal of Advanced Computer Science and Applications</a> </i> &nbsp;
The use of transfer learning made it possible to use a ready-made model that was trained on a parallel corpus of Simple English Wikipedia and not create a simplification corpus in Kazakh from scratch.  ...  For this, a transfer learning technology for simplifying sentences of the Kazakh language has been developed, based on training a neural model for simplifying sentences in the English language.  ...  To assess the quality of the model, we use the BLEU and SARI metrics.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.14569/ijacsa.2021.0120914">doi:10.14569/ijacsa.2021.0120914</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/defel3mdkjaazkudf5cwfeub5e">fatcat:defel3mdkjaazkudf5cwfeub5e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211011213011/https://thesai.org/Downloads/Volume12No9/Paper_14-Development_of_Technology.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/9f/e4/9fe454b99e4c3ebb1bb517b1ff45d9a98c9497f2.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.14569/ijacsa.2021.0120914"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> Publisher / doi.org </button> </a>

Flexible Generation of Natural Language Deductions [article]

Kaj Bostrom, Xinyu Zhao, Swarat Chaudhuri, Greg Durrett
<span title="2021-09-09">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We evaluate our models using out-of-domain sentence compositions from the QASC (Khot et al., 2020) and EntailmentBank (Dalvi et al., 2021) datasets as well as targeted perturbation sets.  ...  Natural language is an attractive representation for this purpose -- it is both highly expressive and easy for humans to understand.  ...  This work was partially supported by NSF awards # IIS-1814522 and # CCF-1918651, by DARPA KAIROS award # FA8750-19-2-1003, and by ARO award # W911NF-21-1-0009.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2104.08825v2">arXiv:2104.08825v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/b4dy552stja33pqqcpahxmkokq">fatcat:b4dy552stja33pqqcpahxmkokq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210915094702/https://arxiv.org/pdf/2104.08825v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/91/1b/911b7539e964782670e555930b291de16fa971c5.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2104.08825v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Automated Paraphrase Lattice Creation for HyTER Machine Translation Evaluation

Marianna Apidianaki, Guillaume Wisniewski, Anne Cocos, Chris Callison-Burch
<span title="">2018</span> <i title="Association for Computational Linguistics"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/d5ex6ucxtrfz3clshlkh3f6w2q" style="color: black;">Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)</a> </i> &nbsp;
The original HyTER metric relied on hand-crafted paraphrase networks which restricted its applicability to new data. We test, for the first time, HyTER with automatically built paraphrase lattices.  ...  We show that although the metric obtains good results on small and carefully curated data with both manually and automatically selected substitutes, it achieves medium performance on much larger and noisier  ...  Evaluating HyTER with Automatic Substitutions We assess the quality of HyTERA to evaluate the quality of MT output both at the sentence and the system level.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/n18-2077">doi:10.18653/v1/n18-2077</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/naacl/ApidianakiWCC18.html">dblp:conf/naacl/ApidianakiWCC18</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/466a3liavbh2fjlxaprkcvy5ly">fatcat:466a3liavbh2fjlxaprkcvy5ly</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190301124240/http://www.aclweb.org:80/anthology/N18-2077" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f1/73/f1730a82163373479a6dd9b2d86953f8d5f68a17.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/n18-2077"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Rethinking Crowd Sourcing for Semantic Similarity [article]

Shaul Solomon and Adam Cohn and Hernan Rosenblum and Chezi Hershkovitz and Ivan P. Yamshchikov
<span title="2021-09-24">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Estimation of semantic similarity is crucial for a variety of natural language processing (NLP) tasks.  ...  The paper offers heuristics to filter out unreliable annotators and stimulates further discussions on human perception of semantic similarity.  ...  The Data To see differences in the semantic tendencies of human annotation, we used several standard paraphrase and style transfer datasets alongside a random selection of sentence pairs from each dataset  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2109.11969v1">arXiv:2109.11969v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4ojmomlfofhazaqnhngfvemwte">fatcat:4ojmomlfofhazaqnhngfvemwte</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210929161706/https://arxiv.org/pdf/2109.11969v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e0/34/e034b1c96c824f55baa39a240a93905e2a4148f8.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2109.11969v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

On the Evaluation of Vision-and-Language Navigation Instructions [article]

Ming Zhao, Peter Anderson, Vihan Jain, Su Wang, Alexander Ku, Jason Baldridge, Eugene Ie
<span title="2021-01-26">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Vision-and-Language Navigation wayfinding agents can be enhanced by exploiting automatically generated navigation instructions.  ...  However, existing instruction generators have not been comprehensively evaluated, and the automatic evaluation metrics used to develop them have not been validated.  ...  Acknowledgements We thank the Google Data Compute team, in particular Ashwin Kakarla and Priyanka Rachapally, for their tooling and annotation support for this project.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.10504v1">arXiv:2101.10504v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hlnosquwh5b2fgllgzzqfqieiu">fatcat:hlnosquwh5b2fgllgzzqfqieiu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210130160526/https://arxiv.org/pdf/2101.10504v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/ca/e0/cae06461bfc6978e9f3a3bbb0b0d0840cd762720.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2101.10504v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Syntax-guided Controlled Generation of Paraphrases [article]

Ashutosh Kumar, Kabir Ahuja, Raghuram Vadapalli, Partha Talukdar
<span title="2020-05-18">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We perform extensive automated and human evaluations over multiple real-world English language datasets to demonstrate the efficacy of SGCP over state-of-the-art baselines.  ...  We address this limitation in the paper and propose Syntax Guided Controlled Paraphraser (SGCP), an end-to-end framework for syntactic paraphrase generation.  ...  We thank the action editor Asli Celikyilmaz, and the three anonymous reviewers for their helpful suggestions in preparing the manuscript.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.08417v1">arXiv:2005.08417v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ymbci3ej5fb2nodzcfjwo7c5qe">fatcat:ymbci3ej5fb2nodzcfjwo7c5qe</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200528233422/https://arxiv.org/pdf/2005.08417v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.08417v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

ViWOZ: A Multi-Domain Task-Oriented Dialogue Systems Dataset For Low-resource Language [article]

Phi Nguyen Van, Tung Cao Hoang, Dung Nguyen Manh, Quan Nguyen Minh, Long Tran Quoc
<span title="2022-03-15">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Furthermore, we provide a comprehensive benchmark of both modular and end-to-end models in low-resource language scenarios.  ...  Therefore, their performance in low-resource languages is still a significant problem due to the absence of a standard dataset and evaluation policy.  ...  Despite the automation on the system side, the quality of those datasets is heavily influenced by the quality of the dialogue system.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.07742v1">arXiv:2203.07742v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6t2j2hv5mndszolerxedxcp2mi">fatcat:6t2j2hv5mndszolerxedxcp2mi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220317004714/https://arxiv.org/pdf/2203.07742v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/88/e5/88e5c6d4112953b5075043f0a44246fea3d4e3d5.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.07742v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Style-transfer and Paraphrase: Looking for a Sensible Semantic Similarity Metric [article]

Ivan P. Yamshchikov, Viacheslav Shibaev, Nikolay Khlebnikov, Alexey Tikhonov
<span title="2020-12-03">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The rapid development of such natural language processing tasks as style transfer, paraphrase, and machine translation often calls for the use of semantic similarity metrics.  ...  Using a new dataset of fourteen thousand sentence pairs human-labeled according to their semantic similarity, we demonstrate that none of the metrics widely used in the literature is close enough to human  ...  Introduction Style transfer and paraphrase are two tasks in Natural Language Processing (NLP). Both of them are centered around the problem of an automated reformulation.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.05001v3">arXiv:2004.05001v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/33m5e5lwbnccpkrx4pelcpx24a">fatcat:33m5e5lwbnccpkrx4pelcpx24a</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200414203010/https://arxiv.org/pdf/2004.05001v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.05001v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

How Well Sentence Embeddings Capture Meaning

Lyndon White, Roberto Togneri, Wei Liu, Mohammed Bennamoun
<span title="">2015</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/ym4qbiso25flxacjl7t5dh6k3i" style="color: black;">Proceedings of the 20th Australasian Document Computing Symposium on ZZZ - ADCS &#39;15</a> </i> &nbsp;
Several existing models, including URAE, PV-DM and PV-DBOW, were assessed against a bag of words benchmark.  ...  Depending on the model used for the embeddings this will vary -different models are suited for different down-stream applications.  ...  Acknowledgement This research is supported by the Australian Postgraduate Award, and partially funded by Australian Research Council DP150102405 and LP110100050.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/2838931.2838932">doi:10.1145/2838931.2838932</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/adcs/WhiteTLB15.html">dblp:conf/adcs/WhiteTLB15</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/b756omor4jbynpdq6dtt4qdlei">fatcat:b756omor4jbynpdq6dtt4qdlei</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20160308101305/http://white.ucc.asn.au/publications/White2015SentVecMeaning.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/72/a2/72a234ba0695c7740e6ee3b8c0952c8e9cd13ee6.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1145/2838931.2838932"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> acm.org </button> </a>

Automatic Question Generation using Sequence to Sequence RNN Model

<span title="2020-03-10">2020</span> <i title="Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/cj3bm7tgcffurfop7xzswxuks4" style="color: black;">VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE</a> </i> &nbsp;
In this research, we suggest using the neural encoder-decoder model to produce substantive and complex questions from the sentences of natural language.  ...  We apply a attention-based sequence to sequence learning paradigm for the task and analyze the impact of encoding sentence vs. knowledge at paragraph level.  ...  Such approaches ' main uses include literacy understanding and language testing which may not suit academic learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.35940/ijitee.e2675.039520">doi:10.35940/ijitee.e2675.039520</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vdjapplfffcllaaz23ntzemlyy">fatcat:vdjapplfffcllaaz23ntzemlyy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220224035539/https://www.ijitee.org/wp-content/uploads/papers/v9i5/E2675039520.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/1c/e3/1ce36492b70ec9b5337e03af79ed9f731f216c8b.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.35940/ijitee.e2675.039520"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

ParCourE: A Parallel Corpus Explorer for a Massively Multilingual Corpus [article]

Ayyoob Imani, Masoud Jalili Sabet, Philipp Dufter, Michael Cysouw, Hinrich Schütze
<span title="2021-07-15">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Examples include assessing language similarity for effective transfer learning, injecting inductive biases into machine learning models or creating resources such as dictionaries and inflection tables.  ...  ParCourE can be set up for any parallel corpus and can thus be used for typological research on other corpora as well as for exploring their quality and properties.  ...  Examples include assessing language similarity for effective transfer learning, injecting inductive biases into machine learning models and creating resources such as dictionaries and inflection tables  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.06632v2">arXiv:2107.06632v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/sq37ij4dfvgptca5bk4omizf44">fatcat:sq37ij4dfvgptca5bk4omizf44</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210721015814/https://arxiv.org/pdf/2107.06632v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/59/c6/59c6400221188cec53a676f980a0dc89120ee119.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2107.06632v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Entailment Relation Aware Paraphrase Generation [article]

Abhilasha Sancheti, Balaji Vasan Srinivasan, Rachel Rudinger
<span title="2022-03-20">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We propose a reinforcement learning-based weakly-supervised paraphrasing system, ERAP, that can be trained using existing paraphrase and natural language inference (NLI) corpora without an explicit task-specific  ...  A combination of automated and human evaluations show that ERAP generates paraphrases conforming to the specified entailment relation and are of good quality as compared to the baselines and uncontrolled  ...  ERAP (Figure 2 ) consists of a paraphrase generator ( §2.1) and an evaluator ( §2.2) comprising of various scorers to assess the quality of generated paraphrases for different aspects.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.10483v1">arXiv:2203.10483v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ezvu22zxgbe5ln5xrothsruhca">fatcat:ezvu22zxgbe5ln5xrothsruhca</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220608165212/https://arxiv.org/pdf/2203.10483v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c4/4a/c44a192fc5bd431f0b30691b92f6c72207c8b8eb.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2203.10483v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Improving Automated Evaluation of Open Domain Dialog via Diverse Reference Augmentation [article]

Varun Gangal, Harsh Jhamtani, Eduard Hovy, Taylor Berg-Kirkpatrick
<span title="2021-06-05">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Prior work has shown the importance of having multiple valid reference responses for meaningful and robust automated evaluations.  ...  More specifically, we use (1) a commonsense knowledge base to elicit a large number of plausible reactions given the dialog history (2) relevant instances retrieved from dialog corpus, using similar past  ...  Acknowledgements We thank anonymous ACL reviewers for insightful comments and feedback. We thank Prakhar Gupta (Gupta et al., 2019) for useful discussions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.02833v1">arXiv:2106.02833v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/46fppdxzfvhexj5yod2qbpk7o4">fatcat:46fppdxzfvhexj5yod2qbpk7o4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210611142929/https://arxiv.org/pdf/2106.02833v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/43/94/43946aa78c80b6d7e6ffff837bdf4cff85f6a935.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.02833v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 3,102 results