NLPIR is one of the key academic conferences to
present research results and new developments in
the area of the Natural Language Processing and
Information Retrieval. For its 9th edition,
NLPIR 2025 will be held in Kyushu University,
Fukuoka, Japan during December 12-14, 2025.
The topics of interests for submission include,
but are not limited to:
- Core NLP & Data Science
- AI-Driven Methods & Innovations
- Cross-Cutting Themes
- Emerging Frontiers
Foundations of Language
Processing
•Data/text mining, corpus
linguistics, and
psycholinguistic modeling
•Basic NLP pipelines:
tokenization, POS tagging,
lemmatization, dependency
parsing, and semantic role
labeling
•Low-resource language
engineering and cross-lingual
adaptation
Linguistic Analysis &
Understanding
•Syntax, semantics, discourse
analysis, and pragmatics
•Multimodal speech
recognition/synthesis (ASR/TTS)
and conversational AI
•Diachronic corpora, temporal
reasoning, and evolving language
models
Knowledge Systems & Semantics
•Automated knowledge
acquisition, ontology
generation/alignment, and
semantic web technologies
•Neuro-symbolic integration:
combining logic-based reasoning
with neural networks
Content Analysis & IR
•Topic modeling, event/anomaly
detection, and sentiment/emotion
analysis
•Document summarization,
plagiarism detection, and
authorship attribution
•Dynamic/personalized IR,
adversarial retrieval, and
cross-language systems
Social & Multimedia Analysis
•Personality/emotion detection
in social media, misinformation
tracking
•Multimodal IR (text, image,
video) and virality prediction
Large Language Models
(LLMs) & Transformers
•Architectures (BERT, GPT, T5, LLaMA) for NLU, generation, and few-shot learning
•Domain-specific LLMs (e.g., BioGPT, Codex) and tools like ChatGPT, DeepSeek,
Claude
•Ethical challenges: bias mitigation, hallucination control, and AI-generated
content detection
Generative AI & Automation
•Abstractive summarization, synthetic data generation, and conversational agents
•Multimodal LLMs (e.g., GPT-4V) for vision-language tasks
Graph & Deep Learning
•GNNs for co-occurrence graphs, knowledge graph completion, and dynamic networks
•Swarm intelligence hybridized with transformer architectures
Efficiency & Scalability
•Model compression (pruning, quantization), federated learning, and edge NLP
•Distributed training frameworks for trillion-parameter models
Human-Centric NLP
•Interactive AI: chatbots, dynamic query resolution, and personalized
recommendation systems
•Explainability (XAI) and visualization of attention mechanisms
Machine Translation &
Multilinguality
•Zero-shot translation, LLM-driven low-resource adaptation, and post-editing
workflows
Decentralized & Collaborative
Systems
•Blockchain for decentralized knowledge graphs, federated search, and
privacy-preserving NLP
Ethics & Governance
•AI safety, fairness audits, and regulatory compliance (e.g., EU AI Act)
•Combatting misinformation and deepfakes in social/content platforms
AI for Science: LLMs
in biomedical NLP, climate text analysis, and legal document processing
Embodied AI: Language
models integrated with robotics and real-world interaction
Self-Supervised Learning:
Pre-training paradigms beyond transformers