Picture

Weiting (Steven) Tan @weiting_nlp

Ph.D. Candidate at @jhuclsp, Student Researcher @Bytedance Seed | Prev @AIatMeta @Amazon Alexa AI steventan0110.github.io USA Joined July 2021
  • Tweets
    76
  • Followers
    210
  • Following
    295
  • Likes
    164
Weiting (Steven) Tan retweeted
sanxing_chen Profile Picture

Sanxing Chen @sanxing_chen

3 days ago

Most RL for LLMs today is single-step optimization on a given state (e.g., an instruction), which is essentially a bandit setup. But to learn a meta-policy that can solve various bandit problems via in-context trial and error, you need true multi-turn RL over a long horizon. So,…

Weiting (Steven) Tan retweeted
jacobaustin132 Profile Picture

Jacob Austin @jacobaustin132

8 months ago

Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n

Weiting (Steven) Tan retweeted
DongfuJiang Profile Picture

Dongfu Jiang @DongfuJiang

a month ago

🚀 Excited to finally share our paper on VerlTool, released today after months of work since the initial release in late May! VerlTool is a high-efficiency, easy-to-use framework for Agentic RL with Tool use (ARLT), built on top of VeRL. It currently supports a wide range of…

DongfuJiang Profile Picture

Dongfu Jiang @DongfuJiang

4 months ago

🚀 Excited to finally share our paper on VerlTool, released today after months of work since the initial release in late May! VerlTool is a high-efficiency, easy-to-use framework for Agentic RL with Tool use (ARLT), built on top of VeRL. It currently supports a wide range of… https://t.co/MK4OAY49Uz

Weiting (Steven) Tan retweeted
jaseweston Profile Picture

Jason Weston @jaseweston

a month ago

🌀Diversity Aware RL (DARLING)🌀 📝: arxiv.org/abs/2509.02534 - Jointly optimizes for quality & diversity using a learned partition function - Outperforms standard RL in quality AND diversity metrics, e.g. higher pass@1/p@k - Works for both non-verifiable & verifiable tasks 🧵1/5

Weiting (Steven) Tan retweeted
ben_vandurme Profile Picture

Benjamin Van Durme @ben_vandurme

7 months ago

Our latest on compressed representations: Key-Value Distillation (KVD). Query-independen transformer compression, with offline supervised distillation.

Weiting (Steven) Tan retweeted
deepseek_ai Profile Picture

DeepSeek @deepseek_ai

9 months ago

🛠️ DeepSeek-R1: Technical Highlights 📈 Large-scale RL in post-training 🏆 Significant performance boost with minimal labeled data 🔢 Math, code, and reasoning tasks on par with OpenAI-o1 📄 More details: github.com/deepseek-ai/De… 🐋 4/n

Weiting (Steven) Tan retweeted
JHUCompSci Profile Picture

JHU Computer Science @JHUCompSci

9 months ago

Congratulations to Prof. Philipp Koehn on being named a Fellow of the @aclmeeting! cs.jhu.edu/news/philipp-k…

weiting_nlp Profile Picture

Weiting (Steven) Tan @weiting_nlp

10 months ago

I had a great time helping host MASC-SLL at Hopkins last year. MASC-SLL is a great opportunity to connect with fellow AI/NLP/Speech researchers. If your organization is in the Mid-Atlantic region and is interested in hosting the event, please reach out!

MASC_Conference Profile Picture

MASC-ALL Conference @MASC_Conference

10 months ago

I had a great time helping host MASC-SLL at Hopkins last year. MASC-SLL is a great opportunity to connect with fellow AI/NLP/Speech researchers. If your organization is in the Mid-Atlantic region and is interested in hosting the event, please reach out!

Weiting (Steven) Tan retweeted
tli104 Profile Picture

Tianjian Li @tli104

10 months ago

I have written a blogpost offering an explanation of why both the chosen and the rejected log-probability decreases during DPO, and more interestingly, why it is a desired phenomenon to some extent. Link: tianjianl.github.io/blog/2024/dpo/

Weiting (Steven) Tan retweeted
sherjilozair Profile Picture

Sherjil Ozair @sherjilozair

10 months ago

Very happy to hear that GANs are getting the test of time award at NeurIPS 2024. The NeurIPS test of time awards are given to papers which have stood the test of the time for a decade. I took some time to reminisce how GANs came about and how AI has evolve in the last decade.

weiting_nlp Profile Picture

Weiting (Steven) Tan @weiting_nlp

12 months ago

Excited to see that SpiritLM is fully open-sourced now. It supports speech and text as both input and output. Please consider trying it at: github.com/facebookresear…

AIatMeta Profile Picture

AI at Meta @AIatMeta

12 months ago

Excited to see that SpiritLM is fully open-sourced now. It supports speech and text as both input and output. Please consider trying it at: github.com/facebookresear…

Weiting (Steven) Tan retweeted
sainingxie Profile Picture

Saining Xie @sainingxie

12 months ago

Representation matters. Representation matters. Representation matters, even for generative models. We might've been training our diffusion models the wrong way this whole time. Meet REPA: Training Diffusion Transformers is easier than you think! sihyun.me/REPA/(🧵1/n)

Weiting (Steven) Tan retweeted
fe1ixxu Profile Picture

Haoran Xu @fe1ixxu

a year ago

Multilingual models are usually heavily skewed in favor of high-resource languages. We change this with X-ALMA: an LLM-based translator committed to ensuring top-tier performance across 50 diverse languages, regardless of their resource levels! Paper: arxiv.org/pdf/2410.03115

Trends for United States

#SmackDown
11,7 B posts
Kyren Lacy
19,1 B posts
Diddy
148 B posts
#WNBAFinals
1.904 posts
End 1Q
Melo
12,6 B posts
End of 1st
1.669 posts
Sami
21,4 B posts
#loveafterlockup
Hamas
822 B posts
Aleister Black
Jay Jones
37,2 B posts
Jewell
#BelleCollective
Only 4
61,6 B posts
#iahsfb
Delaware
11 B posts
Usman
16,1 B posts
Arthur Jones
6.577 posts
After 1
623 B posts

You might like

Final year Ph.D. candidate in NLP, CV @jhuclsp. Researching transparent reasoning, multimodality, and fact verification.
Kate Sanders
@kesnet50
306 Followers
Asst. Prof. in CS at @BU_Tweets ≡ {Mechanistic, causal} {interpretability, NLP} ≡ Formerly: PhD @jhuclsp
Aaron Mueller
@amuuueller
2K Followers
Phd student at @osunlp | Research Intern at AI2 PRIOR @allen_ai | Previous: MS @jhuclsp; Intern @Amazon
Boyuan Zheng@ICML
@boyuan__zheng
768 Followers
PhD student @uwnlp | visiting researcher @AIatMeta | undergrad @jhuclsp #NLProc
Stella Li @COLM2025
@StellaLisy
3K Followers
CS PhD candidate @umdcs. #NLProc at @ClipUmd | Commonsense + xNLP, AI, CompLing | 2x Ex-Research Intern @Microsoft @MSFTResearch
Shramay Palta
@PaltaShramay
849 Followers
PhDing @stanfordnlp & Chilling @FutureHouseSF | teaching language models to do research
CLS
@ChengleiSi
5K Followers
Embracing existential crisis as technological progress 1st Author @ NeurIPS,AAAI,EMNLPx2,NAACL,EACL. Ideas cheap, execution lacking human and compute cycles
Suzanna Sia
@suzyahyah
347 Followers
Research Intern @meta & PhD @jhu, prev @databricks MosaicML @microsoft, @mstranslator, @GeorgiaTech Working on datasets
Marc Marone
@ruyimarone
809 Followers
Senior Researcher @Microsoft | Ph.D. @JHU '24| ex-intern @Microsoft| @Meta AI | @Amazon Alexa AI
Haoran Xu
@fe1ixxu
541 Followers
NLP + AI assistant prof. @UTAustin CS, postdoc @uncnlp w/ @mohitban47, PhD @jhuclsp, @NSF grad fellow. Building communicative+collaborative AI.
Elias Stengel-Eskin
@EliasEskin
2K Followers
CS PhD student @uncnlp @unc, Research Intern @salesforce | prev research intern @meta and @amazon, work on video reasoning , long videos.
Ziyang Wang
@ZiyangW00
359 Followers
@JHUCompSci Research Assistant @jhuclsp @mdredze Research interests: Alignment LLMs, Model Editing, Interpretability. (విజయ్ మురారి)
Vijay Murari Tiyyala
@VijayTiyyala
181 Followers
CS PhD student at @Stanford. Previously undergrad at @jhuclsp
Abe Hou
@abe_hou
413 Followers
Johns Hopkins / Microsoft
Benjamin Van Durme
@ben_vandurme
1K Followers
Trustworthy AI/ML in healthcare & high-stakes apps | My job is (mostly) error bars 🫡 (eg, conformal prediction) | CS PhD at Johns Hopkins. Prev at Yale. he/him
Drew Prinster
@DrewPrinster
333 Followers