zhiyuanpeng/SPTAR
Soft Prompt Tuning for Augmenting Dense Retrieval with Large Language Models
SPTAR helps improve how relevant information is found from large collections of text, like financial news or articles, when you have many search queries but not enough labeled examples to train a search system effectively. It takes an existing collection of documents and a set of queries, and enhances the search system's ability to pull out highly relevant documents, even for complex or nuanced queries. This tool is for professionals such as researchers, data scientists, or information retrieval specialists who need to build high-performing search systems.
No commits in the last 6 months.
Use this if you are working with large document collections and need to significantly improve the accuracy of your search or information retrieval system, especially when you have limited manually labeled query-document pairs.
Not ideal if you already have a perfectly performing retrieval system or if your information retrieval needs are very basic and don't require advanced augmentation techniques.
Stars
16
Forks
6
Language
Jupyter Notebook
License
—
Category
Last pushed
Feb 19, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/zhiyuanpeng/SPTAR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
aalok-sathe/surprisal
A unified interface for computing surprisal (log probabilities) from language models! Supports...
EvolvingLMMs-Lab/lmms-engine
A simple, unified multimodal models training engine. Lean, flexible, and built for hacking at scale.
reasoning-machines/pal
PaL: Program-Aided Language Models (ICML 2023)
FunnySaltyFish/Better-Ruozhiba
【逐条处理完成】人为审核+修改每一条的弱智吧精选问题QA数据集
microsoft/monitors4codegen
Code and Data artifact for NeurIPS 2023 paper - "Monitor-Guided Decoding of Code LMs with Static...