A new research paper proposes a framework to improve document re-ranking by distinguishing between conceptual entity relevance and observable entity relevance. The authors argue that current entity-aware retrieval methods incorrectly assume that topically relevant entities are always useful for ranking. They introduce Observable Entity Relevance (OER) as a measure of whether an entity's observed presence in a collection effectively discriminates relevant from non-relevant documents. Experiments show that aligning supervision with OER significantly improves document pruning and retrieval performance compared to traditional methods like BM25. AI
排序理由 The cluster contains a research paper submitted to arXiv detailing a new framework for document re-ranking.
在 arXiv cs.IR (Information Retrieval) 阅读 →
- alphaXiv
- arXiv
- BM25
- CatalyzeX Code Finder for Papers
- Conceptual Entity Relevance
- CORE Recommender
- DagsHub
- Entity Labels Are Not Entity Signals: A Framework for Observable Relevance in Document Re-Ranking
- Gotit.pub
- Hugging Face
- Influence Flower
- Observable Entity Relevance
- ScienceCast
- Utshab Kumar Ghosh
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →