---

ACL2023読み会@名大

---

概要

• 会場: 名古屋大学工学部IB電子情報館中棟1階 IB015講義室
• 日時: 2023年09月28日(木)
• 形式: 1人18分(発表15分、質疑3分)

---

プログラム

• 09:30-09:35: Opening [笹野遼平, slide]
• 09:35-10:30: Session 1: Large Language Model 1 (座長: 柴克樹)
  • Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM’s Translation Capability [笹野遼平, slide]
  • Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters [高野春樹, slide]
  • Say What You Mean! Large Language Models Speak Too Positively about Negative Commonsense Knowledge [吉田朝飛, slide]
• 10:40-11:35: Session 2: Large Language Model 2 (座長: 柴克樹)
  • Pre-trained Language Models Can be Fully Zero-Shot Learners [大鹿雅史, slide]
  • LAMBADA: Backward Chaining for Automated Reasoning in Natural Language [加藤芳秀]
  • Can Large Language Models Be an Alternative to Human Evaluation? [宮川由衣, slide]
• 11:35-12:30: Lunch Break
• 12:30-13:25: Session 3: Dialogue System (座長: 青野広太郎)
  • One Cannot Stand for Everyone! Leveraging Multiple User Simulators to train Task-oriented Dialogue Systems [伊藤滉一朗, slide]
  • Learning New Skills after Deployment: Improving open-domain internet-driven dialogue with human feedback [刀祢有紀彦, slide]
  • Resolving Indirect Referring Expressions for Entity Selection [徳久良子, slide]
• 13:35-14:30: Session 4: Retrieval (座長: 青野広太郎)
  • SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval [矢野千紘, slide]
  • DataFinder: Scientific Dataset Recommendation from Natural Language Descriptions [生駒流季, slide]
  • Chain-of-Skills: A Configurable Model for Open-Domain Question Answering [杉浦尚弥, slide]
• 14:40-15:35: Session 5: Miscellaneous (座長: 笹野遼平)
  • Weaker Than You Think: A Critical Look at Weakly Supervised Learning [陽田祥平, slide]
  • When and how to paraphrase for named entity recognition? [駒水孝裕, slide]
  • DICE: Data-Efficient Clinical Event Extraction with Generative Models [三輪誠, slide]
• 15:35-15:55: Short Break
• 15:55-16:50: Session 6: Embedding (座長: 唐晨)
  • Interpretable Word Sense Representations via Definition Generation: The Case of Semantic Change Analysis [韓毅, slide]
  • WhitenedCSE: Whitening-based Contrastive Learning of Sentence Embeddings [塚越駿, slide]
  • Shrinking Embeddings for Hyper-Relational Knowledge Graphs [井田龍希, slide]
• 17:00-17:55: Session 7: Dialogue Dataset (座長: 唐晨)
  • FIREBALL: A Dataset of Dungeons and Dragons Actual-Play with Structured Game State Information [周旭琳, slide]
  • VSTAR: A Video-grounded Dialogue Dataset for Situated Semantic Understanding with Scene and Topic Transitions [市川拓茉, slide]
  • Annotating and Detecting Fine-grained Factual Errors for Dialogue Summarization [小川泰弘, slide]
• 17:55-18:00: Closing

---

発表者

• 笹野遼平, 韓毅, 塚越駿, 高野春樹, 矢野千紘, 大鹿雅史, 陽田祥平 (武田・笹野研), 加藤芳秀, 生駒流季, 伊藤滉一朗, 吉田朝飛(松原研), 周旭琳, 市川拓茉, 刀祢有紀彦(東中研), 駒水孝裕, 宮川由衣(井手・駒水研), 杉浦尚弥(外山研), 三輪誠, 井田龍希(豊田工大), 小川泰弘(名古屋市大), 徳久良子(豊田中研/東北大)

---

過去の読み会

• ACL2019読み会@名大
• ACL2018読み会@名大

---