Researchers have developed Dhara, a novel retrieval engine designed to efficiently search Indian Knowledge Systems (IKS). This system integrates hybrid search, entity notes, fuzzy verse retrieval, and neural reranking to locate relevant evidence that may be scattered across various texts, including verses, translations, and commentaries. Evaluations show that Dhara's BGE-M3 embedding model outperforms traditional BM25, and the inclusion of entity notes and fuzzy verse retrieval further enhances retrieval accuracy across different categories. AI
IMPACT This system could improve access to and understanding of historical and cultural knowledge bases through advanced AI-driven search.
RANK_REASON The cluster describes a new research paper detailing a novel information retrieval system. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- Avinash Holenarasipur Nagendra
- BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
- BM25
- ColBERTv2
- Gemini
- Hugging Face
- Indian Knowledge Systems
- Qwen
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →