Resources and Evaluation Information Extraction Dialogue and Interactive Systems Efficient Methods for NLP Language Modeling and Analysis of LMs Machine Learning for NLP Natural Language Generation Question Answering Interpretability, Interactivity and Analy… Information Retrieval and Text Mining Machine Translation Sentiment Analysis, Stylistic Analysis,… Language Grounding to Vision, Robotics… Semantics: Lexical, Sentence level, Docu… Speech and Multimodality Summarization Computational Social Science and Cultu… Multilinguality and Linguistic Diversity Ethics in NLP Commonsense Reasoning 0 200 400 600 研究分野の統計 5 Submitted Accepted (Main + Findings) 応⽤研究が最多
Resources and Evaluation Information Extraction Dialogue and Interactive Systems Efficient Methods for NLP Language Modeling and Analysis of LMs Machine Learning for NLP Natural Language Generation Question Answering Interpretability, Interactivity and Analy… Information Retrieval and Text Mining Machine Translation Sentiment Analysis, Stylistic Analysis,… Language Grounding to Vision, Robotics… Semantics: Lexical, Sentence level, Docu… Speech and Multimodality Summarization Computational Social Science and Cultu… Multilinguality and Linguistic Diversity Ethics in NLP Commonsense Reasoning 0 200 400 600 研究分野の統計 6 Submitted Accepted (Main + Findings) 次点で,トレンドのLLM (テーマトラック)
Resources and Evaluation Information Extraction Dialogue and Interactive Systems Efficient Methods for NLP Language Modeling and Analysis of LMs Machine Learning for NLP Natural Language Generation Question Answering Interpretability, Interactivity and Analy… Information Retrieval and Text Mining Machine Translation Sentiment Analysis, Stylistic Analysis,… Language Grounding to Vision, Robotics… Semantics: Lexical, Sentence level, Docu… Speech and Multimodality Summarization Computational Social Science and Cultu… Multilinguality and Linguistic Diversity Ethics in NLP Commonsense Reasoning 0 200 400 600 研究分野の統計 7 Submitted Accepted (Main + Findings) • リソース・評価系 • 情報抽出系 • 対話系 が続く
in the Age of LLMs: Nothing but blue skies! ü “⼤企業が圧倒的研究規模で幅を利かせるのは昔から同じ.今 LLM 時代は始ま ったばかりであり,取り組むべき課題・魅⼒的な機会は沢⼭ある” ü ⼩規模でも画期的な研究ができる,ということを研究事例を紹介しつつ指南 Emily Mower Provost, University of Michigan From Speech to Emotion to Mood: Mental Health Modeling in Real- World Environments ü ⾳声感情認識による双極性障害の⾃動診断システムに関する講演 ü 「実環境適⽤における課題(堅牢性,汎⽤性)に直⾯」から「継続学習活⽤によ る⽇常⾳声からの⾼精度な感情認識を実現」までの⻑期にわたる取組みを紹介 Jong Park, Korea Advanced Institute of Science and Technology Human-Centric Natural Language Processing ⾔語的なマイノリティ(聴覚障害者,精神障害者など)のためのNLP研究につい て,最近の成果を多数紹介.特に,⼿話コーパスの構築など低資源ドメインの資 源整備を推進している点が印象的だった.