[논문 간단 정리] Advancing continual lifelong learning in neural information retrieval: definition, dataset, framework, and empirical evaluation

Advancing continual lifelong learning in neural information retrieval: definition, dataset, framework, and empirical evaluation

• Publication Info: Information Sciences 2025
• URL: https://www.sciencedirect.com/science/article/pii/S0020025524012829

---

Contribution

• IR task의 맥락에서 Continual Learning 패러다임에 대해 명확히 정의함
• Continual IR을 평가하기 위해, Topic-MS-MARCO 데이터셋을 제안함
  • 주제별 IR task와 predefined task similarity가 포함
• CLNIR (Continual Learning Framework for Neural Information Retrieval) 프레임워크 제안
  • regularization & replay mechanism에 기반한 7가지 전략

Limitations

• optimization-based & architecture-based strategies는 탐구되지 않음
• 단일 데이터셋만을 사용했음
  • cross-domain setting이라고 보기 힘듦