Jingtao Zhan

DRhard. SIGIR'21: Optimizing DR with hard negatives and achieving SOTA first-stage retrieval performance on TREC DL Track.

RepCONC. WSDM'22 Best Paper: Learning Discrete Representations via Constrained Clustering for Effective and Efficient Dense Retrieval

RepBERT-Index. RepBERT is a competitive first-stage retrieval technique. It represents documents and queries with fixed-length contextualized embeddings. The inner products of them are regarded as relevance scores. Its efficiency is comparable to bag-of-words methods.

disentangled-retriever. An easy-to-use python toolkit for flexibly adapting various neural ranking models to target domain.

JPQ. CIKM'21: JPQ substantially improves the efficiency of Dense Retrieval with 30x compression ratio, 10x CPU speedup and 2x GPU speedup.

IntelligenceTest. An evaluation framework to test AI in a trial-and-error process. It is a simplified Natural Selection test.

bert-ranking-analysis. SIGIR'20: An Analysis of BERT in Document Ranking

PromptReformulate. Python

extrapolate-eval. CIKM 2022: Evaluating Interpolation and Extrapolation Performance of Neural Retrieval Models

DRScale. Python

pyserini. Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

guide. The Student's Guide to @lintool

BERT-related-papers. BERT-related papers

warc-clueweb. Python library for reading ClueWeb09's warc files

DL4EEG-Classification. The implementation of deep learning models for EEG classification.

jingtaozhan.

Rayuela.jl. Code for my PhD thesis. Library of quantization-based methods for fast similarity search in high dimensions. Presented at ECCV 18.