MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human Retrievers
Jushaan Kalra, Xinran Zhao, To Eun Kim, Fengyu Cai, Fernando Diaz, Tongshuang Wu · 2025
Read MoR: Better Handling Diverse Queries with a Mixture of Sparse, Dense, and Human Retrievers by Jushaan Kalra, Xinran Zhao, To Eun Kim, Fengyu Cai, Fernando Diaz, Tongshuang Wu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-15862.txt.
Description
Jushaan Kalra and five co-workers wrote a 2025 paper on search tools for AI systems that answer with sources. Some tools match words, others match meaning, and people can search too. Each gives a distinct, useful signal. In practice, teams pick one tool by rule of thumb, which fails on varied questions.
Their mixture picks and weights several tools for each question, with no manual choice. The mix is small but beat each single tool and even larger models in tests. Adding human search as one tool in the mix helped a lot. The work shows that varied tools work better as a team.
Catalog record
| CREATOR | Jushaan Kalra, Xinran Zhao, To Eun Kim, Fengyu Cai, Fernando Diaz, Tongshuang Wu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-15862