NAACL
#NLP#LanguageModel#Bias#PostTraining#PerplexityCurse
Issue Date: 2025-05-02 Where is the answer? Investigating Positional Bias in Language Model Knowledge Extraction, Kuniaki Saito+, NAACL25 Comment元ポスト:https://x.com/losnuevetoros/status/1918332232181207096?s=46&t=Y6UuIHB0Lv0IpmFAjlc2-Q#SimulST(SimultaneousSpeechTranslation)
Issue Date: 2025-04-30 Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translation, Sara Papi+, NAACL22 Comment同時翻訳研究で主要なmetricの一つ関連:#1915 ... #DocumentSummarization#NLP#Dataset
Issue Date: 2018-06-29 Newsroom: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies, Max+, NAACL18 Comment文書要約に使用可能なデータセット 38の出版元からデータを収集し、サイズは1.3M article程度 既存のデータセットと比較すると、Coverageが高く生成的なものを多く含むことが特徴 詳細は：https://summari.es ... #NeuralNetwork#Sentence#Embeddings#NLP
Issue Date: 2017-12-28 Learning Distributed Representations of Sentences from Unlabelled Data, Hill+, NAACL16 CommentSentenceのrepresentationを学習する話 代表的なsentenceのrepresentation作成手法(CBOW, SkipGram, SkipThought, Paragraph Vec, NMTなど)をsupervisedな評価（タスク志向+supervised）とun ... #NeuralNetwork#InformationRetrieval#SearchEngine#MultitaskLearning#QueryClassification#WebSearch
Issue Date: 2018-02-05 Representation Learning Using Multi-Task Deep Neural Networks for Semantic Classification and Information Retrieval, Liu+, NAACL-HLT15 Commentクエリ分類と検索をNeural Netを用いてmulti-task learningする研究分類(multi-class classification)とランキング(pairwise learning-to-rank)という異なる操作が必要なタスクを、multi task learningの枠組みで ... #NaturalLanguageGeneration#SingleFramework#NLP#ConceptToTextGeneration
Issue Date: 2017-12-31 Unsupervised concept-to-text generation with hypergraphs, Konstas+, NAACL-HLT12 #NaturalLanguageGeneration#DataDriven#NLP#ConceptToTextGeneration
Issue Date: 2017-12-31 Aggregation via set partitioning for natural language generation, Barzilay+, HLT-NAACL06 #DocumentSummarization#Document#NLP#Admin'sPick
Issue Date: 2018-01-21 Cut and paste based text summarization, Jing+, NAACL00 CommentAbstractiveなSummarizationの先駆け的研究。 AbstractiveなSummarizationを研究するなら、押さえておいたほうが良い。 ... #Article#NeuralNetwork#NaturalLanguageGeneration#NLP#DataToTextGeneration
Issue Date: 2017-12-31 What to talk about and how? Selective Generation using LSTMs with Coarse-to-Fine Alignment, Mei+, NAACL-HLT’16 Commentcontent-selectionとsurface realizationをencoder-decoder alignerを用いて同時に解いたという話。 普通のAttention basedなモデルにRefinerとPre-Selectorと呼ばれる機構を追加。通常のattentionにはatte ... #Article#DocumentSummarization#NLP#IntegerLinearProgramming (ILP)#Update
Issue Date: 2017-12-28 Improving Update Summarization via Supervised ILP and Sentence Reranking, Li et al. NAACL’15, 2015.05 Comment・update summarizationをILPで定式化．基本的なMDSのILPのterm weightingにsalienceの要素に加えてnoveltyの要素を加える．term weightingにはbigramを用いる．bigram使うとよくなることがupdate summarization ... #Article#Multi#PersonalizedDocumentSummarization#NLP#SearchEngine
Issue Date: 2017-12-28 WebInEssence: A Personalized Web-Based Multi-Document Summarization and Recommendation System, Radev+, NAACL, 01, 2001.06 Comment・ドキュメントはオフラインでクラスタリングされており，各クラスタごとにmulti-document summarizationを行うことで， ユーザが最も興味のあるクラスタを同定することに役立てる．あるいは検索結果のページのドキュメントの要約を行う． 要約した結果には，extractした文の元U ... #Article#PersonalizedDocumentSummarization#NLP#Admin'sPick
Issue Date: 2017-12-28 A Study for Documents Summarization based on Personal Annotation, HLT-NAACL-DUC’03, Zhang+, 2003, 2003.05 Comment #SimulST(SimultaneousSpeechTranslation)
Issue Date: 2025-04-30 Over-Generation Cannot Be Rewarded: Length-Adaptive Average Lagging for Simultaneous Speech Translation, Sara Papi+, NAACL22 Comment同時翻訳研究で主要なmetricの一つ関連:#1915 ... #DocumentSummarization#NLP#Dataset
Issue Date: 2018-06-29 Newsroom: A Dataset of 1.3 Million Summaries with Diverse Extractive Strategies, Max+, NAACL18 Comment文書要約に使用可能なデータセット 38の出版元からデータを収集し、サイズは1.3M article程度 既存のデータセットと比較すると、Coverageが高く生成的なものを多く含むことが特徴 詳細は：https://summari.es ... #NeuralNetwork#Sentence#Embeddings#NLP
Issue Date: 2017-12-28 Learning Distributed Representations of Sentences from Unlabelled Data, Hill+, NAACL16 CommentSentenceのrepresentationを学習する話 代表的なsentenceのrepresentation作成手法(CBOW, SkipGram, SkipThought, Paragraph Vec, NMTなど)をsupervisedな評価（タスク志向+supervised）とun ... #NeuralNetwork#InformationRetrieval#SearchEngine#MultitaskLearning#QueryClassification#WebSearch
Issue Date: 2018-02-05 Representation Learning Using Multi-Task Deep Neural Networks for Semantic Classification and Information Retrieval, Liu+, NAACL-HLT15 Commentクエリ分類と検索をNeural Netを用いてmulti-task learningする研究分類(multi-class classification)とランキング(pairwise learning-to-rank)という異なる操作が必要なタスクを、multi task learningの枠組みで ... #NaturalLanguageGeneration#SingleFramework#NLP#ConceptToTextGeneration
Issue Date: 2017-12-31 Unsupervised concept-to-text generation with hypergraphs, Konstas+, NAACL-HLT12 #NaturalLanguageGeneration#DataDriven#NLP#ConceptToTextGeneration
Issue Date: 2017-12-31 Aggregation via set partitioning for natural language generation, Barzilay+, HLT-NAACL06 #DocumentSummarization#Document#NLP#Admin'sPick
Issue Date: 2018-01-21 Cut and paste based text summarization, Jing+, NAACL00 CommentAbstractiveなSummarizationの先駆け的研究。 AbstractiveなSummarizationを研究するなら、押さえておいたほうが良い。 ... #Article#NeuralNetwork#NaturalLanguageGeneration#NLP#DataToTextGeneration
Issue Date: 2017-12-31 What to talk about and how? Selective Generation using LSTMs with Coarse-to-Fine Alignment, Mei+, NAACL-HLT’16 Commentcontent-selectionとsurface realizationをencoder-decoder alignerを用いて同時に解いたという話。 普通のAttention basedなモデルにRefinerとPre-Selectorと呼ばれる機構を追加。通常のattentionにはatte ... #Article#DocumentSummarization#NLP#IntegerLinearProgramming (ILP)#Update
Issue Date: 2017-12-28 Improving Update Summarization via Supervised ILP and Sentence Reranking, Li et al. NAACL’15, 2015.05 Comment・update summarizationをILPで定式化．基本的なMDSのILPのterm weightingにsalienceの要素に加えてnoveltyの要素を加える．term weightingにはbigramを用いる．bigram使うとよくなることがupdate summarization ... #Article#Multi#PersonalizedDocumentSummarization#NLP#SearchEngine
Issue Date: 2017-12-28 WebInEssence: A Personalized Web-Based Multi-Document Summarization and Recommendation System, Radev+, NAACL, 01, 2001.06 Comment・ドキュメントはオフラインでクラスタリングされており，各クラスタごとにmulti-document summarizationを行うことで， ユーザが最も興味のあるクラスタを同定することに役立てる．あるいは検索結果のページのドキュメントの要約を行う． 要約した結果には，extractした文の元U ... #Article#PersonalizedDocumentSummarization#NLP#Admin'sPick
Issue Date: 2017-12-28 A Study for Documents Summarization based on Personal Annotation, HLT-NAACL-DUC’03, Zhang+, 2003, 2003.05 Comment ![image](https://user-images.githubuse重 ...