Corrections: February 2023 (#2377)

* Author name correction for 2022.nllp-1.29, closes #2374.

* Author name correction for 2022.findings-emnlp.483, closes #2372.

* Author name correction for 2022.emnlp-main.696, closes #2371.

* Paper pdf correction for 2022.emnlp-main.313, closes #2370.

* Paper pdf correction for 2022.mrl-1.9, closes #2365.

* Added 2022.gem-1 to siggen, closes #2367.

* Author name metadata correction for 2022.findings-emnlp.75, closes #2378.

* Paper pdf correction for 2022.emnlp-main.435, closes #2383.

* Author order correction for 2022.findings-emnlp.183, closes #2382.

* updated full proceedings and ingested one missing paper for TSAR-22 proceedings, closes #2369.

* ingested one missing paper for TSAR-22 proceedings.

* Abstract metadata correction for 2022.eurali-1.9, closes #2387.

* Author name metadata correction for W19-5606, closes #2351.

* ingested missing paper for 2022.inlg-main, updated main proceedings, closes #2324.

* Mattia Di Gangi

* Paper pdf correction for 2022.findings-emnlp.185, closes #2390.

* Paper pdf correction for 2021.scil-1.14, closes #2389.

* Paper pdf correction for 2021.scil-1.19, closes #2391.

* Paper pdf correction for 2022.wanlp, closes #2373.

* Author name metadata correction for 2022.findings-acl.177, closes #2183.

* Paper pdf correction for 2022.conll-1.3, closes #2394.

* Paper pdf correction for 2022.wmt-1.44, closes #2397.

* Author name meta data correction for 2022.blackboxnlp-1.25, closes #2396.

* Author name correction for 2022.lrec-1.617, closes #2402.

* Author name correction for 2022.sigdial-1.14, closes #2400.

* Paper pdf correction for 2022.wmt-1.54, closes #2403.

* Paper pdf correction for 2022.sustainlp-1.11, closes #2295.

* replaced pdf for W19-6805, closes #1889.

* Added sig info for 2022.sigdial-1, closes #2408.

Co-authored-by: Daniel Gildea <gildea>

data/xml/2021.scil.xml

@@ -130,8 +130,10 @@
       <title>Information flow, artificial phonology and typology</title>
       <author><first>Adamantios</first><last>Gafos</last></author>
       <pages>148–157</pages>
-      <url hash="e5193f77">2021.scil-1.14</url>
+      <url hash="9b763950">2021.scil-1.14</url>
       <bibkey>gafos-2021-information</bibkey>
+      <revision id="1" href="2021.scil-1.14v1" hash="e5193f77"/>
+      <revision id="2" href="2021.scil-1.14v2" hash="9b763950" date="2023-02-15">Added references and corrected typos.</revision>
     </paper>
     <paper id="15">
       <title>Learnability of indexed constraint analyses of phonological opacity</title>
@@ -169,8 +171,10 @@
       <author><first>Brandon</first><last>Waldon</last></author>
       <author><first>Judith</first><last>Degen</last></author>
       <pages>206–215</pages>
-      <url hash="6404bfe2">2021.scil-1.19</url>
+      <url hash="6e1e3a41">2021.scil-1.19</url>
       <bibkey>waldon-degen-2021-modeling</bibkey>
+      <revision id="1" href="2021.scil-1.19v1" hash="6404bfe2"/>
+      <revision id="2" href="2021.scil-1.19v2" hash="6e1e3a41" date="2023-02-15">Corrected typos.</revision>
     </paper>
     <paper id="20">
       <title>Multiple alignments of inflectional paradigms</title>

data/xml/2022.blackboxnlp.xml

@@ -278,7 +278,7 @@
       <title>Garden Path Traversal in <fixed-case>GPT</fixed-case>-2</title>
       <author><first>William</first><last>Jurayj</last></author>
       <author><first>William</first><last>Rudman</last></author>
-      <author><first>Carsten</first><last>Eickhof</last></author>
+      <author><first>Carsten</first><last>Eickhoff</last></author>
       <pages>305-313</pages>
       <abstract>In recent years, large-scale transformer decoders such as the GPT-x family of models have become increasingly popular. Studies examining the behavior of these models tend to focus only on the output of the language modeling head and avoid analysis of the internal states of the transformer decoder. In this study, we present a collection of methods to analyze the hidden states of GPT-2 and use the model’s navigation of garden path sentences as a case study. To enable this, we compile the largest currently available dataset of garden path sentences. We show that Manhattan distances and cosine similarities provide more reliable insights compared to established surprisal methods that analyze next-token probabilities computed by a language modeling head. Using these methods, we find that negating tokens have minimal impacts on the model’s representations for unambiguous forms of sentences with ambiguity solely over what the object of a verb is, but have a more substantial impact of representations for unambiguous sentences whose ambiguity would stem from the voice of a verb. Further, we find that analyzing the decoder model’s hidden states reveals periods of ambiguity that might conclude in a garden path effect but happen not to, whereas surprisal analyses routinely miss this detail.</abstract>
       <url hash="725e7562">2022.blackboxnlp-1.25</url>

data/xml/2022.conll.xml

@@ -44,8 +44,10 @@
       <author><first>Kathleen</first><last>Carley</last></author>
       <pages>27-39</pages>
       <abstract>This paper investigates how hate speech varies in systematic ways according to the identities it targets. Across multiple hate speech datasets annotated for targeted identities, we find that classifiers trained on hate speech targeting specific identity groups struggle to generalize to other targeted identities. This provides empirical evidence for differences in hate speech by target identity; we then investigate which patterns structure this variation. We find that the targeted demographic category (e.g. gender/sexuality or race/ethnicity) appears to have a greater effect on the language of hate speech than does the relative social power of the targeted identity group. We also find that words associated with hate speech targeting specific identities often relate to stereotypes, histories of oppression, current social movements, and other social contexts specific to identities. These experiments suggest the importance of considering targeted identity, as well as the social contexts associated with these identities, in automated hate speech classification</abstract>
-      <url hash="4448b7e0">2022.conll-1.3</url>
+      <url hash="9eddbf11">2022.conll-1.3</url>
       <bibkey>yoder-etal-2022-hate</bibkey>
+      <revision id="1" href="2022.conll-1.3v1" hash="4448b7e0"/>
+      <revision id="2" href="2022.conll-1.3v2" hash="9eddbf11" date="2023-02-16">Updated wording.</revision>
     </paper>
     <paper id="4">
       <title>Continual Learning for Natural Language Generations with Transformer Calibration</title>

data/xml/2022.eamt.xml

@@ -39,7 +39,7 @@
     </paper>
     <paper id="2">
       <title>Neural Speech Translation: From Neural Machine Translation to Direct Speech Translation</title>
-      <author><first>Mattia Antonino Di</first><last>Gangi</last></author>
+      <author><first>Mattia Antonino</first><last>Di Gangi</last></author>
       <pages>7–8</pages>
       <url hash="e7f82669">2022.eamt-1.2</url>
       <bibkey>gangi-2022-neural</bibkey>
@@ -783,7 +783,7 @@
     </paper>
     <paper id="65">
       <title>Automatic Video Dubbing at <fixed-case>A</fixed-case>pp<fixed-case>T</fixed-case>ek</title>
-      <author><first>Mattia Di</first><last>Gangi</last></author>
+      <author><first>Mattia</first><last>Di Gangi</last></author>
       <author><first>Nick</first><last>Rossenbach</last></author>
       <author><first>Alejandro</first><last>Pérez</last></author>
       <author><first>Parnia</first><last>Bahar</last></author>

data/xml/2022.emnlp.xml

@@ -3775,8 +3775,10 @@
       <author><first>Yi</first><last>Chang</last></author>
       <pages>4746-4758</pages>
       <abstract>Weakly-supervised text classification aims to train a classifier using only class descriptions and unlabeled data. Recent research shows that keyword-driven methods can achieve state-of-the-art performance on various tasks. However, these methods not only rely on carefully-crafted class descriptions to obtain class-specific keywords but also require substantial amount of unlabeled data and takes a long time to train. This paper proposes FastClass, an efficient weakly-supervised classification approach. It uses dense text representation to retrieve class-relevant documents from external unlabeled corpus and selects an optimal subset to train a classifier. Compared to keyword-driven methods, our approach is less reliant on initial class descriptions as it no longer needs to expand each class description into a set of class-specific keywords.Experiments on a wide range of classification tasks show that the proposed approach frequently outperforms keyword-driven models in terms of classification accuracy and often enjoys orders-of-magnitude faster training speed.</abstract>
-      <url hash="beb61fee">2022.emnlp-main.313</url>
+      <url hash="d523c7e1">2022.emnlp-main.313</url>
       <bibkey>xia-etal-2022-fastclass</bibkey>
+      <revision id="1" href="2022.emnlp-main.313v1" hash="beb61fee"/>
+      <revision id="2" href="2022.emnlp-main.313v2" hash="d523c7e1" date="2023-02-07">Corrected author name.</revision>
     </paper>
     <paper id="314">
       <title>Neural-Symbolic Inference for Robust Autoregressive Graph Parsing via Compositional Uncertainty Quantification</title>
@@ -5284,8 +5286,10 @@
       <author><first>Ruifeng</first><last>Xu</last></author>
       <pages>6485-6498</pages>
       <abstract>Aspect Sentiment Triplet Extraction (ASTE) aims to extract the aspect terms along with the corresponding opinion terms and the expressed sentiments in the review, which is an important task in sentiment analysis. Previous research efforts generally address the ASTE task in an end-to-end fashion through the table-filling formalization, in which the triplets are represented by a two-dimensional (2D) table of word-pair relations. Under this formalization, a term-level relation is decomposed into multiple independent word-level relations, which leads to relation inconsistency and boundary insensitivity in the face of multi-word aspect terms and opinion terms. To overcome these issues, we propose Boundary-Driven Table-Filling (BDTF), which represents each triplet as a relation region in the 2D table and transforms the ASTE task into detection and classification of relation regions. We also notice that the quality of the table representation greatly affects the performance of BDTF. Therefore, we develop an effective relation representation learning approach to learn the table representation, which can fully exploit both word-to-word interactions and relation-to-relation interactions. Experiments on several public benchmarks show that the proposed approach achieves state-of-the-art performances.</abstract>
-      <url hash="41d0dcdb">2022.emnlp-main.435</url>
+      <url hash="e659525b">2022.emnlp-main.435</url>
       <bibkey>zhang-etal-2022-boundary</bibkey>
+      <revision id="1" href="2022.emnlp-main.435v1" hash="41d0dcdb"/>
+      <revision id="2" href="2022.emnlp-main.435v2" hash="e659525b" date="2023-02-10">Corrected Figure 2.</revision>
     </paper>
     <paper id="436">
       <title>Attention and Edge-Label Guided Graph Convolutional Networks for Named Entity Recognition</title>
@@ -8396,9 +8400,9 @@
     </paper>
     <paper id="696">
       <title>Dictionary-Assisted Supervised Contrastive Learning</title>
-      <author><first>Patrick</first><last>Wu</last></author>
+      <author><first>Patrick Y.</first><last>Wu</last></author>
       <author><first>Richard</first><last>Bonneau</last></author>
-      <author><first>Joshua</first><last>Tucker</last></author>
+      <author><first>Joshua A.</first><last>Tucker</last></author>
       <author><first>Jonathan</first><last>Nagler</last></author>
       <pages>10217-10235</pages>
       <abstract>Text analysis in the social sciences often involves using specialized dictionaries to reason with abstract concepts, such as perceptions about the economy or abuse on social media. These dictionaries allow researchers to impart domain knowledge and note subtle usages of words relating to a concept(s) of interest. We introduce the dictionary-assisted supervised contrastive learning (DASCL) objective, allowing researchers to leverage specialized dictionaries when fine-tuning pretrained language models. The text is first keyword simplified: a common, fixed token replaces any word in the corpus that appears in the dictionary(ies) relevant to the concept of interest. During fine-tuning, a supervised contrastive objective draws closer the embeddings of the original and keyword-simplified texts of the same class while pushing further apart the embeddings of different classes. The keyword-simplified texts of the same class are more textually similar than their original text counterparts, which additionally draws the embeddings of the same class closer together. Combining DASCL and cross-entropy improves classification performance metrics in few-shot learning settings and social science applications compared to using cross-entropy alone and alternative contrastive and data augmentation methods.</abstract>

data/xml/2022.eurali.xml

@@ -111,7 +111,7 @@
       <author><first>Maksim</first><last>Melenchenko</last></author>
       <author><first>Dmitry</first><last>Novokshanov</last></author>
       <pages>61–64</pages>
-      <abstract>This poster describes the Shughni Documentation Project consisting of the Online Shughni Dictionary, morphological analyzer, orthography converter, and Shughni corpus. The online dictionary has not only basic functions such as finding words but also facilitates more complex tasks. Representing a lexeme as a network of database sections makes it possible to search in particular domains (e.g., in meanings only), and the system of labels facilitates conditional search queries. Apart from this, users can make search queries and view entries in different orthographies of the Shughni language and send feedback in case they spot mistakes. Editors can add, modify, or delete entries without programming skills via an intuitive interface. In future, such website architecture can be applied to creating a lexical database of Iranian languages. The morphological analyzer performs automatic analysis of Shughni texts, which is useful for linguistic research and documentation. Once the analysis is complete, homonymy resolution must be conducted so that the annotated texts are ready to be uploaded to the corpus. The analyzer makes use of the orthographic converter, which helps to tackle the problem of spelling variability in Shughni, a language with no standard literary tradition.</abstract>
+      <abstract>This paper describes the Shughni Documentation Project consisting of the Online Shughni Dictionary, morphological analyzer, orthography converter, and Shughni corpus. The online dictionary has not only basic functions such as finding words but also facilitates more complex tasks. Representing a lexeme as a network of database sections makes it possible to search in particular domains (e.g., in meanings only), and the system of labels facilitates conditional search queries. Apart from this, users can make search queries and view entries in different orthographies of the Shughni language and send feedback in case they spot mistakes. Editors can add, modify, or delete entries without programming skills via an intuitive interface. In future, such website architecture can be applied to creating a lexical database of Iranian languages. The morphological analyzer performs automatic analysis of Shughni texts, which is useful for linguistic research and documentation. Once the analysis is complete, homonymy resolution must be conducted so that the annotated texts are ready to be uploaded to the corpus. The analyzer makes use of the orthographic converter, which helps to tackle the problem of spelling variability in Shughni, a language with no standard literary tradition.</abstract>
       <url hash="568091b6">2022.eurali-1.9</url>
       <bibkey>makarov-etal-2022-digital</bibkey>
     </paper>