ML, NLP, Quant research. Graduate from @ntunlpsg | @AstarI2r | @uoftcompsci | @centralesupelec. Currently QR at ADIA.
United Arab Emirates
Joined August 2014
- Tweets190
- Following2.6K
- Followers424
- Likes2.1K
Checkout our analysis paper, to be featured in ACL main conference:
arxiv.org/abs/2310.10570
We investigate the "middle-curse" exposed by the LITM paper for the specific task of abstractive summarizarion. In brief: the middle curse is very much an issue ! w @JotyShafiq @AixinSG โ at Central Region, Singapore
Checkout our paper on SSMs for long-input summarization! Happening today at EACL ๐
Congrats to @florian_flb @songdng
๐ค Ever wondered how to summarize an entire book ๐๐๐ ๐๐ก ๐๐๐๐? We delve into this question in our recent paper โ๐๐๐๐๐๐: ๐๐ญ๐๐ญ๐-๐๐ฉ๐๐๐ ๐๐จ๐๐๐ฅ๐ฌ ๐๐จ๐ซ ๐๐จ๐ง๐ ๐๐จ๐๐ฎ๐ฆ๐๐ง๐ญ ๐๐๐ฌ๐ญ๐ซ๐๐๐ญ๐ข๐ฏ๐ ๐๐ฎ๐ฆ๐ฆ๐๐ซ๐ข๐ณ๐๐ญ๐ข๐จ๐งโ (#EACL2024). ๐งต (1/8)
5
304
Ever wondered how to summarize an entire book *all at once* with a deep learning model?
Check out our latest paper on summarization:
arxiv.org/abs/2401.17919
Through State Space models, we can scale the input to 600k tokens (!)
1
7
390
Check our survey - ChatGPT is gradually getting dethroned by open source alternatives
It has been exactly one year since the release of ChatGPT. How far are open-source LLMs? We provide an exhaustive review of open-source LLMs that claim to catch up with or surpass ChatGPT in various capabilities.
paper๐: arxiv.org/pdf/2311.16989.pdf
๐งต(1/5)
6
310
Checkout our latest work on summarization!
Working paper: LLMs prioritize the introductory section of lengthy documents when generating summaries.
arxiv.org/abs/2310.10570
A related question: what do LLMs really gain from an extremely long context? Is there some form of "summarization" when taking in the long context?
Not happy about the abstractive summaries that you are generating from the language model?
Have a try at sampling many and re-ranking them! ๐
arxiv.org/abs/2212.09593 โ at Central Region, Singapore
In our new ACL Findings 2023 paper from @ntunlpsg , we propose to re rank summary candidates (similarly to models like SimCLS, SummaReranker, BRIO, etc) but this time in an *unsupervised* manner. โ at Central Region, Singapore
1
53
20 years later, Musk re doing a X company. I doubt eBay will buy the company this time though ๐ โ at West Region, Singapore
1
225
Check out the newest survey from our lab! @ntunlpsg โ at Central Region, Singapore
With Palm-E and GPT4, Multimodal models are gaining popularity. To improve their grounding and correctness, retrieval augmentation is a promising solution... Check out our survey on methods that explore the intersection! arxiv.org/abs/2303.10868
1
208
Singapore remains underrated ๐ โ at Central Region, Singapore
This quoted post is unavailable.
4
1,749
~6 years doing research and ~2 years into my PhD, it feels good to reach this 200 citations milestone ๐ฅณ to the moon!๐ โ at West Region, Singapore
3
81
Check out our new paper ! ๐ฅ
To be featured at EMNLP 2022
#emnlp2022 #NLProc โ at West Region, Singapore
Towards Summary Candidates Fusion
deepai.org/publication/towarโฆ
by @MatRavox et al. including @JotyShafiq
#Computation #Language
Honoured to have presented our work at ACL in Dublin today @aclmeeting โ at Dublin City, Ireland
3
14
Thrilled to share that our paper "SummaReranker: a Multi-Task Re-Ranking Framework for Abstractive Summarization" has been accepted to #acl2022nlp @JotyShafiq @ntunlpsg @nancyfchen1.
(1/10) Excited to share our paper in @JAMANetworkOpen where we leveraged routinely collected data to predict risk of developing diabetes in the population *long in the future* with machine learning.
jamanetwork.com/journals/jamโฆ
I highly recommend working with @VinyasHarish , a rising star in ML for healthcare!
Proud to have been part of this team effort between @Layer6AI @PopHlthAnalytic @ICESOntario and @uoftmedicine! Huge kudos to @MatRavox for leading the effort ๐คฉ
3
(1/13) Thrilled to share my first first-author paper in @Nature_NPJ Digital Medicine on using ML to predict diabetes complications at the population level w @Layer6AI @PopHlthAnalytic :
nature.com/articles/s41746-0โฆ
Boyan Slat's problem solving initiatives are so inspiring (and way underrated).
Hereโs the full presentation on how we hope to turn our ocean plastic catch from a problem into a solution:
youtu.be/hzeI0kth22E
1