PACLIC 39 – Conference papers

• Cover and Front Matter
  Vietnam Institute for Advanced Study in Mathematics; Hanoi, Vietnam; 5-7 December 2025
  Emmanuele Chersoni, Jong-Bok Kim (Editors)
  pp. 1-2
  
• Foreword and Committees
  Emmanuele Chersoni, Le Minh Nguyen, Rachel Edita Oñate Roxas, Shirley N. Dita
  pp. 3-5
  
• Improving Interpretability of Lexical Semantic Change with Neurobiological Features
  Kohei Oda, Hiroya Takamura, Kiyoaki Shirai, Natthawut Kertkeidkachorn
  pp. 6-19
  
• Same Spelling, Different Functions Mah: Evaluating Language Models’ Understanding of Singlish Particles
  Chan Young Jung
  pp. 20-30
  
• Evaluating Syntactic Generalization in Transformer-Based Models Using Korean Honorific Agreement
  Nayoung Kwon, Seongmin Mun
  pp. 31-42
  
• Towards Improving Low-Resource Machine Translation with Lightweight Training and Synthetic Data: Case Study of Vietnamese - Khmer
  Trong Huy Nguyen, Thanh Huong Le, Que Nhu Tran
  pp. 43-55
  
• Researching and Rethinking the Culture and Practice of Eldercare in Hong Kong: A Corpus-assisted Discourse Analysis
  Jesse W. C. Yip, Michelle Man-Long Pang
  pp. 56-68
  
• Syllabic Distribution and Developmental Patterns of Mandarin Glides in Preschool Children
  I-Ping Wan, Xiang Li, Yu Ching Tsai
  pp. 69-83
  
• Assessing GPT Models’ Sensitivity to Epistemic Meanings in Korean Periphrastic Construction
  Yebin Lee, Arum Kang, Sanghoun Song
  pp. 84-93
  
• Speech-Like Cues and the Limits of Musicality: Lexical Tone Normalization in Mandarin across Speech, Rap, and Song Contexts
  Yujia Tian, Yanyuan Ye, Mency Lu, JIA Fanlu, Ran Tao
  pp. 94-101
  
• RMCP: Enhancing LLM-based Translation via Prompting with Retrieved Monolingual Corpora Wenhui Shi and Ziman Han Institute of Language Sciences Shanghai International Studies University
  Wenhui Shi, Ziman Han
  pp. 102-117
  
• Preserving Ambiguity: Prompt Sensitivity in Gender-Neutral Literary Translation by GPT Models
  Jin Yim
  pp. 118-130
  
• Bridging the Modality Gap by Similarity Standardization with Pseudo-Positive Samples
  Shuhei Yamashita, Daiki Shirafuji, Tatsuhiko Saito
  pp. 131-144
  
• PRISM: A Pedagogical Multi-Agent for Structured Group Learning
  Van-Khanh Tran, Van-Khai Dang, Duc-Huy Nguyen
  pp. 145-154
  
• A Persona Dialogue Dataset of Lesser-Known Characters for Fairer Evaluation of Role-Playing LLMs
  Ryuichi Uehara, Michimasa Inaba
  pp. 155-168
  
• Exaggeration Scoring of News Summaries through LLM-based Relative Judgments
  Keisuke Iwamoto, Kazutaka Shimada
  pp. 169-178
  
• BV-FRD: A Multimodal Vietnamese-English Food Review Video Description Generation
  Bao Pham-Thai, Vy Do Le Khanh, Huy Quoc To
  pp. 179-194
  
• Large-Scale Japanese Metaphor Corpus Construction: Expanding BCCWJ-Metaphor with Automated Annotation
  Hang Zhu, Rowan Hall Maudslay, Kanako Komiya, Sachi Kato, Masayuki Asahara
  pp. 195-204
  
• Converging and diverging variations in metaphorization of light verbs: A corpus study on four Chinese speech communities
  Ka-Fai Yip, Yaxuan Ji, Dicky Ko, Thomas Ho, and Benjamin K. Tsou
  pp. 205-219
  
• Effect of Emotional Congruency and Cognitive Load on Word Processing
  Jieyu Chen, Yujia Tian, Jing Qi, Ran Tao
  pp. 220-230
  
• Toward True Neutrality: Evaluating Inference-Time Debiasing Strategies for Gender Coreference Resolution in LLMs
  Arati Mohapatra, S Jaya Nirmala
  pp. 231-240
  
• A/E Pelita as an Aspectual Marker in Korean Auxiliary Verb Constructions: An Experimental Comparison with Spanish Optionalse
  Nakyung Yoon
  pp. 241-249
  
• ‘But this one was so . . . male.’ A Corpus-Based and LLM-Augmented Analysis of Language and Gender Bias in Barbie
  Xin Luo, Wing-hei Lok, Yuyin-Hsu
  pp. 250-258
  
• A corpus-assisted metaphor analysis in portraying businesswomen: Diachronic changes in Hong Kong English news
  Yifan Li, Yanlin Li, Kathleen Ahrens
  pp. 259-267
  
• Analysis of the Correlation Between Theory of Mind and Dialogue Ability to Identify Essential ToM for Dialogue Systems
  Haruhisa Iseno, Atsumoto Ohashi, Tetsuji Ogawa, Shinnosuke Takamichi, Ryuichiro Higashinaka
  pp. 268-279
  
• An Empirical Survey of Model Merging Algorithms for Social Bias Mitigation
  Daiki Shirafuji, Tatsuhiko Saito, Yasutomo Kimura
  pp. 280-300
  
• The Propositional Idea Densities of Different Languages in Multi-Lingual Parallel Corpus
  Yuka Kaise, Yuto Tsuchiya, Masanori Oya
  pp. 301-305
  
• Scaffolded AI Feedback for L2 Writing: Fostering Self-Correction in Japanese University Students
  Atsushi Nakanishi
  pp. 306-313
  
• ScheduleMe: Multi-Agent Calendar Assistant
  Oshadha Wijerathne, Amandi Nimasha, Dushan Fernando, Nisansa de Silva, and Srinath Perera
  pp. 314-324
  
• AsRED: Development and Evaluation of an Assamese Reduplication Dataset
  Pankaj Choudhury, Chaitanya Kirti, Dhrubajyoti Pathak, Sukumar Nandi
  pp. 325-333
  
• Non-Interactive Symbolic-Aided Chain-of-Thought for Logical Reasoning
  Phuong Minh Nguyen, Tien Huu Dang, Naoya Inoue
  pp. 334-345
  
• Sensory and Affective Dimensions in Mandarin Monosyllabic Adjectives
  Yimei Shao, Yu-Yin Hsu, Chu-Ren Huang
  pp. 346-356
  
• KWordinaryVQA: A Keyword-Driven Generative Visual Question Answering System for Culinary Exploration
  Huy Trieu, Thanh Thai Nguyen, Thanh Nghia Vo, Thinh Vuong Vo, Thanh Tu Dang, Tung Le
  pp. 357-371
  
• MoE4EDiReF: Mixture of Experts for Emotion Discovery and Reasoning its Flip in Conversation
  Katarzyna Szczepaniak, Piotr Andruszkiewicz
  pp. 372-381
  
• Diagnosing the Cultural Gap in Machine Translation of Classical Chinese: A Computational Analysis of The Analects
  Menghan Dong, Youzhi Wu, Siqi Wang, Zhen Wu
  pp. 382-388
  
• From Span Extraction to Classification: A Multi-step Framework for Cognitive Distortion Analysis
  Manh-Cuong Phan, Thi-Ngoc-Phuong Nguyen, Huu-Loi Le, Huy-The Vu, Hajime Hotta, Minh-Tien Nguyen
  pp. 389-398
  
• Is CCGbank Semantically Valid? Insights from Negation Scope Analysis
  Kentaro Kojima, Yoshihide Kato, Shigeki Matsubara
  pp. 399-406
  
• VIDAI : VIDukathAI Interpretation Through Analysis of In-context Reasoning in Tamil using LLMs
  R S Mughil Srinivasan, Kesavan T, Abhijith Balan, Abhinav P M, Parameswari Krishnamurthy, Oswald C
  pp. 407-417
  
• Do Multimodal Large Language Models Have Good Taste? An Evaluation of MLLMs via the Visual Aesthetic Sensitivity Test
  Yi Yao, Zhuang Qiu
  pp. 418-423
  
• L3Cube-MahaEmotions: A Marathi Emotion Recognition Dataset with Synthetic Annotations using CoTR prompting and Large Language Models
  Nidhi Kowtal, Raviraj Joshi
  pp. 424-432
  
• Exploring Synonymy Representation in Large Language Models: A Comparative Analysis with Referential and Use-Based Lexical Resources
  Sara Besharati
  pp. 433-442
  
• Non-English Code Generation with Cross-Lingual Chain of Thought
  Yuha Nishigata, Waka Ito, Kimio Kuramitsu
  pp. 443-451
  
• The Relationship Between Dialogue Acts and Idea Generation in Human–Human Collaborative Story Writing
  Natsumi Ezure, Michimasa Inaba
  pp. 452-465
  
• Non Idiomatic Conventionalised Expressions: A New Pain in the Neck?
  Ganesh Katrapati, Manish Shrivastava
  pp. 466-471
  
• Forecasting Time Series with LLMs via Patch-Based Prompting and Decomposition
  Mayank Bumb, Anshul Vemulapalli, Sri Harsha Jella, Anish Gupta, An La, Ryan Rossi, Franck Dernoncourt, Hongjie Chen, Nesreen Ahmed, Yu Wang
  pp. 472-484
  
• Keeping LLMs from Being Distracted: Grade-Aware Kanji Reading Estimation Fully Executable in Web Browsers for Japanese Education
  Yo Ehara
  pp. 485-494
  
• Controlling Emotion Intensity and Blending in Text via Task Vector Composition for Dialog System Personalization
  Ryota Amano, Kazuya Mera, Yoshiaki Kurosawa, Toshiyuki Takezawa
  pp. 495-504
  
• USA Model: Japanese Universal Sentiment Analysis Model & Construction of Japanese Sentiment Text Classification and Part of Speech Dataset
  Chengguang Gan, Qinghao Zhang, Tatsunori Mori
  pp. 505-515
  
• SiDiaC : Sinhala Diachronic Corpus
  Nevidu Jayatilleke, Nisansa de Silva
  pp. 516-532
  
• The Evolutionary Mechanisms of Transitivization in Mandarin VO Compounds: A Corpus-Driven Study of Competing Alternations
  Menghan Jiang, Chu-ren Huang China
  pp. 533-539
  
• HisGraphRAG: GraphRAG for Vietnamese Historical question answering
  Hoang-Thanh Nguyen, Tung Le, Huy Tien Nguyen
  pp. 540-548
  
• Time Tells: Temporal Event Ordering in Frontier LLMs — Performance, Limitations, and Human Comparison
  Feifei Sun, Ziyi Tong, Teeradaj Racharak, Minh Le Nguyen
  pp. 549-560
  
• Semantic Meaning or Script Shape? A Comparative Study of Cross-Lingual Transfer in mBERT and PIXEL
  Zhenming Li, Kazutaka Shimada
  pp. 561-569
  
• Reference Points in LLM Sentiment Analysis: The Role of Structured Context
  Junichiro Niimi
  pp. 570-580
  
• RoSRL: Adaptive Rule-of-Sum Reinforcement Learning for Efficient and Reliable Summarization
  Thu Phuong Tran Thi, Vinh Nguyen Van, Thai Nguyen Phuong, Quang Vu Ngoc, Khoa Nguyen Dang
  pp. 581-596
  
• Online Information Extraction System (EAOS) for Social Media Comments
  Khanh-Hung Huynh, Phuc Bui Hoang Gia, Huyen Dinh Doan My, Phi-Long Nguyen, Thien-Vu Nguyen-Ho, Trong-Hop Do
  pp. 597-604
  
• On the Role of Contextual Information and Ego States in LLM Agent Behavior for Transactional Analysis Dialogues
  Monika Zamojska, Jarosław A. Chudziak
  pp. 605-613
  
• Enhancing Scientific Title Generation via Optimized Sentence Ordering
  Thanh-Thien Khuu, Thien-Thuan Huynh, Nam Van Chi, Tung Le
  pp. 614-623
  
• Dependency-Aware Word Prediction Integrated with Incremental Parsing
  Hiroki Unno, Tomohiro Ohno, Koichiro Ito, Shigeki Matsubara
  pp. 624-633
  
• How “empirical” is corpus linguistics?: A meta-analysis using formal concept analysis
  Kazuho Kambara, Yuki Sugawara, Norihisa Takahashi
  pp. 634-644
  
• MahaParaphrase: A Marathi Paraphrase Detection Corpus and BERT-based Models
  Suramya Jadhav, Abhay Shanbhag, Amogh Thakurdesai, Ridhima Sinare, Ananya Joshi, Raviraj Joshi
  pp. 645-653
  
• Computational Linguistic Approach to Empathy and its Language Communication Pattern
  Xinyi Wang, Mingyu Wan, Chu-Ren Huang
  pp. 654-661
  
• Ancient Tamil Palm Leaf character recognition using Transformers
  Aswinkumar I, Sangeetha Sivanesan, S Jaya Nirmala
  pp. 662-670
  
• MTikGuard System: A Transformer-Based Multimodal System for Child-Safe Content Moderation on TikTok
  Dat Thanh Nguyen, Nguyen Hung Lam, Anh Thi-Hoang Nguyen, Trong-Hop Do
  pp. 671-682
  
• Personalized Graph-Based Retrieval for Large Language Models
  Steven Au, Cameron J. Dimacali, Ojasmitha Pedirappagari, Namyong Park, Franck Dernoncourt, Yu Wang, Nikos Kanakaris, Hanieh Deilamsalehy, Ryan A. Rossi, Nesreen K. Ahmed
  pp. 683-703
  
• Improving Classical Language Machine Translation using Supervised Fine- T uning with Philological Commentary
  Yuzuki Tsukagoshi, Ikki Ohmukai
  pp. 704-714
  
• Lore Coherent Encounter Generation for Dungeons and Dragons: LLM Fine-tuning and Benchmark
  Aravinth Sivaganeshan, Nisansa de Silva, Akila Peiris
  pp. 715-727
  
• Verb Phrase Idioms in Valency Alternation: A Selection-based Approach
  Tomokazu Takehisa
  pp. 728-736
  
• Improving the Efficiency of Long Document Classification using Sentence Ranking Approach
  Prathamesh Kokate, Mitali Sarnaik, Manavi Khopade, and Raviraj Joshi
  pp. 737-746
  
• Diverse Target Representations for Language Models with Word Difference Representations
  DongNyeong Heo, Daniela Noemi Rim, Heeyoul Choi
  pp. 747-759
  
• How Is Context Important in Named Entity Recognition? A Comparison of Non-contextual and Contextual Word Embeddings
  Dawid Smalcuga, Piotr Andruszkiewicz
  pp. 760-768
  
• IndicSQuAD : A Comprehensive Multilingual Question Answering Dataset for Indic Languages
  Sharvi Endait, Ruturaj Ghatage, Aditya Kulkarni, Rajlaxmi Patil, Raviraj Joshi
  pp. 769-776
  
• HILIGAY NER : A Baseline Named Entity Recognition Model for Hiligaynon
  James Ald Teves, Ray Daniel Cal, Josh Magdiel Villaluz, Jean Malolos, Mico Magtira, Ramon Rodriguez, Joseph Marvin Imperial, Mideth Abisado
  pp. 777-784
  
• An Investigation of Sentiment Polarity in Chinese VO Idioms Based on NLP Models
  Xueyi Wen, Hongzhi Xu, Lili Yang
  pp. 785-795
  
• Domain Adaptation for Multi-document Summarisation: A Case Study in the Medical Research Domain
  Kushan Hewapathirana, Nisansa de Silva C.D. Athuraliya, Piumi Kandanaarachchi
  pp. 796-807
  
• Speaking to the Inner Child: A Discourse Analysis of Healing Narratives in Digital Therapeutic Culture
  Romeo Jr. E. Tejada, Jose Marie E. Ocdenaria, Kenneth C. Lisbo
  pp. 808-817
  
• Dual-mode N-gram Similarity Detection for Forensic Authorship Analysis
  John Blake, Kazuma Tamura, Krzysztof Kredens
  pp. 818-825
  
• A Social Listening System for Beauty Products Using Aspect-Based Sentiment Analysis
  Thanh-Nhi Nguyen, Trong-Hop Do
  pp. 826-834
  
• Balancing Heat and Clarity: Alcohol, Tea, and the Body in Premodern Chinese Texts
  Clio Luo
  pp. 835
  
• From Tea to Symbol: A Multimodal Discourse Analysis of HeyTea's Branding Discourses
  Tian Gao
  pp. 836
  
• Foodies also Need Their Own GPTs: Evaluating Language Models in Visual Question Answering on the WorldCuisines Dataset
  Genta Indra Winata, Emmanuele Chersoni
  pp. 837
  
• Conceptual Metaphors in Food Reviews: LLM-Based Implications for Korean Discourse
  Charmgil Hong, Jong-Bok Kim, Seulkee Park, Yunseong Choe
  pp. 838-840
  
• Landscapes of Matcha: Cultural Semiotics in Japan and Taiwan
  Miwa Morishita, Yasunari Harada
  pp. 841
  
• Toward the Development of a Japanese Food Culture QA
  Waka Ito, Manaka Odagaki, Haruka Tsuchida, Yuha Nishigata, Yui Obara, Kimio Kuramitsu
  pp. 842