Licheng Yu

Research Scientist @Meta, working on Multimodal LLM and media generation.

[Anu Aakash @anuaakash]

swarmed by thoughts ~ meta ai

[Licheng Yu @Licheng_Yu]

@skynetislov3 @MetaAI Sorry we cannot release the model due to the strict data privacy at Meta. The whole work was implemented and deployed using MMF https://t.co/OBmJvCyCrd. I will be very happy to provide more implementation details if you are interested in it. Feel free to ping me.

[Licheng Yu @Licheng_Yu]

@MetaAI Accepted to KDD 2022: https://t.co/OsYFDrrmhc 👏 It's no doubt a great team effort! Looking forward to discussing more at the conference.😄

[AI at Meta @AIatMeta]

Our new multimodal model, CommerceMM, creates rich representations of commerce data. With state-of-the-art performance in product recognition & retrieval, CommerceMM helps tailor suggestions & search results to connect shoppers with the items they want. https://t.co/224JF4RbDM

[Licheng Yu @Licheng_Yu]

@zhoubolei @UCLAengineering @CS_UCLA Congrats!

[Jie Lei @jayleicn]

We have officially announced our #VALUE benchmark for multitask video&language evaluation. It has 11 datasets on 3 video tasks (retrieval, QA, captioning) over diverse domains. VALUE challenge will be hosted at #ICCV2021, join us!

https://t.co/bqzRYZ4LFz
https://t.co/mHHqjfZxG8

[Licheng Yu @Licheng_Yu]

@mohitban47 @unccs @unccollege @UNCResearch @uncnlp Wow! Congratulations! Feeling proud working with you!

[Licheng Yu @Licheng_Yu]

@xinw_ai Congrats!

[Licheng Yu @Licheng_Yu]

@yukez 😅haha..

[Mohit Bansal @mohitban47]

New TVR+TVC datasets on multimodal Video+Dialogue Moment-Retrieval & Captioning!😀

Datasets+code+leaderboards all set up at: https://t.co/AP1s88b9Ci

Come participate and pls feel free to send us questions/suggestions at [email protected]

Pdf: https://t.co/LgXXlZHN6n

[Jie Lei @jayleicn]

Releasing TVR & TVC! Large multimodal video+dialogue moment-retrieval & captioning datasets (109K queries on 22K videos & 262K captions w/ 108K moments); XML+MMT models

Data+Leaderboards+Code: https://t.co/3awMCTkaan

Pdf: https://t.co/5aEMvvi3mS

@Licheng_Yu, TBerg, @MohitBan47

[DeepAI @DeepAI]

TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval
https://t.co/7h6YkrR5jt
by Jie Lei et al. including @mohitban47
#Moment #Computation

[Licheng Yu @Licheng_Yu]

Our new work (also my first work after joining
@Microsoft) "UNITER: LEARNING UNIVERSAL IMAGE-TEXT REPRESENTATIONS"https://t.co/g23xJJi9Oj SOTA on 9 datasets (VQA, VCR, NLVR, Img-Txt Retrieval, Vis-Entailment, Grounding). GREAT effort by everyone, especially Yen-Chun, @LINJIEFUN

[Gabriel Ilharco @gabriel_ilharco]

New state of the art on nine Vision+Language tasks, including VQA, VCR, Visual Entailment, and NLVR2!

Browsing ICLR papers is always a treat

https://t.co/moq5tslNvU

[Mohit Bansal @mohitban47]

TVQA+ v2 is out (paper+data), with 311K annotated bounding boxes for compositional+localization videoQA! Check it out & participate on our leaderboard! Suggestions welcome: [email protected] 😀

Paper: https://t.co/JSWXdKHver
Data/leaderboards: https://t.co/7J4jMHtsjJ

[Mohit Bansal @mohitban47]

We also have a TVQA leaderboard set up at: https://t.co/f2G8D9hjSV -- come participate and pls feel free to send suggestions at [email protected]! 😀

Details in paper: https://t.co/m5YiOR9LLN

#EMNLP2018 #TVQA

[Licheng Yu @Licheng_Yu]

@AlexGuoHan @mohitban47 Thanks!

[Licheng Yu @Licheng_Yu]

Accepted by CVPR2018😀 Code: https://t.co/dn3IYRk6oK https://t.co/t3nWPhOPDp

[Licheng Yu @Licheng_Yu]

Our new work "MAttNet: Modular Attention Network for Referring Expression Comprehension" by @mohitban47 @unccs @AdobeResearch
Paper: https://t.co/6CqNwGggUp
Demo: https://t.co/YJzcdhC51Y

[Licheng Yu @Licheng_Yu]

Our new work "MAttNet: Modular Attention Network for Referring Expression Comprehension" by @mohitban47 @unccs @AdobeResearch
Paper: https://t.co/6CqNwGggUp
Demo: https://t.co/YJzcdhC51Y

[Ken Sakurada @sakuDken]

@kotymgさんのコード使われてた．
“Tutorial: Annotating images with bounding boxes using Amazon Mechanical Turk” by @amazonmturk https://t.co/SKrOCoGBHJ