Shiguang Wu (@shiguang_wu) 's Twitter Profile
Shiguang Wu

@shiguang_wu

MSc student in the IRLAB at Shandong University. Working on Information Retrieval.

ID: 1559960108682989568

linkhttps://furyton.github.io/ calendar_today17-08-2022 17:47:42

19 Tweet

89 Followers

349 Following

Zhaochun Ren (@zhaochun_ren) 's Twitter Profile Photo

Glad to share that our paper "Generative Retrieval as Multi-Vector Dense Retrieval" has been accepted at #SIGIR2024 as a full paper, joint work w/ Shiguang Wu, Pengjie Ren, and Maarten de Rijke

Sumit (@_reachsumit) 's Twitter Profile Photo

Generative Retrieval as Multi-Vector Dense Retrieval Establishes that generative retrieval and multi-vector dense retrieval share a common framework for computing query-document relevance. 📝arxiv.org/abs/2404.00684 👨🏽‍💻github.com/Furyton/GR-as-…

Generative Retrieval as Multi-Vector Dense Retrieval

Establishes that generative retrieval and multi-vector dense retrieval share a common framework for computing query-document relevance.

📝arxiv.org/abs/2404.00684
👨🏽‍💻github.com/Furyton/GR-as-…
Sumit (@_reachsumit) 's Twitter Profile Photo

ExcluIR: Exclusionary Neural Information Retrieval Presents a benchmark and dataset for exclusionary retrieval where queries express what information to exclude. 📝arxiv.org/abs/2404.17288 👨🏽‍💻github.com/zwh-sdu/ExcluIR

ExcluIR: Exclusionary Neural Information Retrieval

Presents a benchmark and dataset for exclusionary retrieval where queries express what information to exclude.

📝arxiv.org/abs/2404.17288
👨🏽‍💻github.com/zwh-sdu/ExcluIR
Shiguang Wu (@shiguang_wu) 's Twitter Profile Photo

The pre-print version of our SIGIR'24 paper on Generative Retrieval as Multi-Vector Dense Retrieval is at arxiv.org/abs/2404.00684. And the code is now available at github.com/Furyton/GR-as-…. We welcome your feedback! Please feel free to leave any comments on our work :) #SIGIR2024

Shiguang Wu (@shiguang_wu) 's Twitter Profile Photo

A significant amount of research has been devoted to analyzing Large Language Models and I've made a paper collection on this topic: github.com/Furyton/awesom…. It covers theoretical/empirical analysis of transformers-based LMs. Your contributions and thoughts are warmly welcomed :)

Iadh Ounis (@iadh) 's Twitter Profile Photo

A very good set of resources (reading list, code, slides, etc) in the webpage of the #sigir2024 Generative Information Retrieval tutorial by Maarten de Rijke et al. - see generative-ir.github.io

A very good set of resources (reading list, code, slides, etc) in the webpage of the  #sigir2024 Generative Information Retrieval tutorial by <a href="/mdr/">Maarten de Rijke</a> et al. - see generative-ir.github.io
Zhaochun Ren (@zhaochun_ren) 's Twitter Profile Photo

Thrilled to share that our paper won the BEST PAPER honorable mention award at #SIGIR2024. Shiguang Wu will present our work today during the Dense Retrieval 2 session (Room Federal A & B, 10h30 to 12h30).

Thrilled to share that our paper won the BEST PAPER honorable mention award at #SIGIR2024. <a href="/shiguang_wu/">Shiguang Wu</a> will present our work today during the Dense Retrieval 2 session (Room Federal A &amp; B, 10h30 to 12h30).
Sumit (@_reachsumit) 's Twitter Profile Photo

Replication and Exploration of Generative Retrieval over Dynamic Corpora Reveals text-based docids outperform numeric-based docids in dynamic corpora, and semantic alignment, fine-grained design, and lexical diversity are key factors for generalization 📝arxiv.org/abs/2504.17519