KSE-Web: An Analysis of Hybrid Retrieval and LLM-Assisted Query Expansion for Low-Resource Khmer Semantic Search

Not specified in the provided content

semantic searchlow-resource languagesquery expansionretrieval methods

2608.21365

Builder Relevance

70%

2h ago

Abstract

This paper presents KSE-Web, an analysis of hybrid retrieval and LLM-assisted query expansion for Khmer semantic search, addressing challenges in low-resource language retrieval.

Reality Card

Core Claim

BM25 retrieval method achieves the strongest overall performance for Khmer semantic search, outperforming LLM-assisted query expansion.

Method / Result

BM25 achieves 0.943 Recall and 0.876 nDCG.

Limitations

LLM-assisted query expansion can introduce topic drift and noisy reformulations, affecting the quality of results.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.

← Back to all papers