Danica Sutherland (@d_j_sutherland) - โปรไฟล์ Twitter

coming back to x, the everything app, to say: Submit work, sign up to review, come to the workshop! This is a great chance to bring together a lot of really cool work that's been happening, but not all as connected (nor as easy to publish) as it should be! testing.ml

Feng Liu@AlexFengLiu1

Excited to share our ICML 2026 Hypothesis Testing Workshop in Seoul, this July! @icmlconf 🎉This workshop aims to bring together researchers developing modern hypothesis testing methodology and applying it to machine learning problems such as robustness, distribution shift, security, medicine, and LLM evaluation. In other words, if you care about how we make ML claims rigorous, this workshop is for you. We now have four confirmed speakers: Arthur Gretton @ArthurGretton, Yao Xie @yaoxie21851119, Bo Li @uiuc_aisecure, and Yisong Yue @yisongyue. The organizing team includes Xiuyuan Cheng (Duke), Feng Liu @AlexFengLiu1, Lester Mackey @LesterMackey, Shayak Sen @shayaksen, Danica J. Sutherland @d_j_sutherland, and Nathaniel Xu (UBC). 📌 Submission deadline: 10 May 2026 📌 Notification: 26 May 2026 📌 Camera-ready: 17 June 2026 📌 Workshop date: July 10 or 11, 2026 (TBA) 🚩Check more information below! 🔗Website: testing.ml 🔗Submission Portal: openreview.net/group?id=ICML.… We’re also recruiting PC members/reviewers. 🔗 Reviewer interest form: docs.google.com/forms/d/e/1FAI… 🏁Please feel free to share this with colleagues, collaborators, and students who may be interested. #ICML #ICML26

English

0

2

3

352

Danica Sutherland รีทวีตแล้ว

Feng Liu@AlexFengLiu1·16 Nis

Excited to share our ICML 2026 Hypothesis Testing Workshop in Seoul, this July! @icmlconf 🎉This workshop aims to bring together researchers developing modern hypothesis testing methodology and applying it to machine learning problems such as robustness, distribution shift, security, medicine, and LLM evaluation. In other words, if you care about how we make ML claims rigorous, this workshop is for you. We now have four confirmed speakers: Arthur Gretton @ArthurGretton, Yao Xie @yaoxie21851119, Bo Li @uiuc_aisecure, and Yisong Yue @yisongyue. The organizing team includes Xiuyuan Cheng (Duke), Feng Liu @AlexFengLiu1, Lester Mackey @LesterMackey, Shayak Sen @shayaksen, Danica J. Sutherland @d_j_sutherland, and Nathaniel Xu (UBC). 📌 Submission deadline: 10 May 2026 📌 Notification: 26 May 2026 📌 Camera-ready: 17 June 2026 📌 Workshop date: July 10 or 11, 2026 (TBA) 🚩Check more information below! 🔗Website: testing.ml 🔗Submission Portal: openreview.net/group?id=ICML.… We’re also recruiting PC members/reviewers. 🔗 Reviewer interest form: docs.google.com/forms/d/e/1FAI… 🏁Please feel free to share this with colleagues, collaborators, and students who may be interested. #ICML #ICML26

English

1

10

55

12.9K

Danica Sutherland รีทวีตแล้ว

Yi (Joshua) Ren@JoshuaRenyi·18 Nis

📢Curious why your LLM behaves strangely after long SFT or DPO? We offer a fresh perspective—consider doing a "force analysis" on your model’s behavior. Check out our #ICLR2025 Oral paper: Learning Dynamics of LLM Finetuning! (0/12)

English

6

115

794

87.4K

Danica Sutherland รีทวีตแล้ว

Ameya Velingker | अमेय वेलिंगकर@ameya_pa·13 Ara

In our Spexphormer method, a smaller network estimates the attention scores of a larger one. This led us to a fundamental question: How small can the network be while still producing accurate estimates? We tackled this question through rigorous theoretical analysis. 1/

English

1

10

1.5K

Danica Sutherland รีทวีตแล้ว

Hamed Shirzad@HamedShirzad13·12 Ara

As a reminder, we will have our poster session tomorrow: 📍 East Exhibit Hall, Poster #3010 📄 arxiv.org/abs/2411.16278 💻 github.com/hamed1375/Sp_E… To motivate you further, we have some insights gained from the attention score analysis of this work, which I'll share in this thread:

Hamed Shirzad@HamedShirzad13

Graph Transformers (GTs) can handle long-range dependencies and resolve information bottlenecks, but they’re computationally expensive. Our new model, Spexphormer, helps scale them to much larger graphs – check it out at @NeurIPSConf next week, or the preview here! #NeurIPS2024

English

1

2

12

1.1K

Danica Sutherland รีทวีตแล้ว

Yi (Joshua) Ren@JoshuaRenyi·9 Ara

LLM's self-play is ubiquitous. What will happen if M[t] iteratively learns from M[t-1] for too many generations? Come and chat with us at NeurIPS 🗓️Friday, Dec 13 📍East Exhibit Hall A-C, Poster #3305 ⏰11:00 AM–2:00 PM PST. [1/7]

English

1

3

11

958

Danica Sutherland รีทวีตแล้ว

Hamed Shirzad@HamedShirzad13·5 Ara

Graph Transformers (GTs) can handle long-range dependencies and resolve information bottlenecks, but they’re computationally expensive. Our new model, Spexphormer, helps scale them to much larger graphs – check it out at @NeurIPSConf next week, or the preview here! #NeurIPS2024

English

1

3

24

3.3K

Danica Sutherland@d_j_sutherland·19 Eyl

@matloff @thegautamkamath n^{-1/3} is better than n^{-1/4}, so I agree that research like that would probably not be of much real value

English

1

0

4

125

Norm Matloff 你有冇諗清楚呀?@matloff·18 Eyl

@thegautamkamath Maybe so, but too much research is of the nature "We improved the old O(n^{-1/3}) rate to O(n^{-1/4})," of no real value to Microsoft or the world at large.

English

1

0

5

1.3K

Gautam Kamath@thegautamkamath·18 Eyl

10 years ago, Microsoft closed down the entire Silicon Valley research lab overnight, almost all members out of a job Microsoft Research seems to have learned better since, but still 1 of the blackest marks on their legacy. A warning to any researcher in industry at any company

Ittai Abraham@ittaia

Happy 10 year anniversary to MSR SVC ❤️ msrsvc.info windowsontheory.org/2014/09/19/far…

English

5

8

195

32K

Danica Sutherland@d_j_sutherland·25 Tem

Grokking on modular arithmetic: early (kernel) phase can overfit but cannot generalize, but with small regularization GD eventually escapes the kernel regime and can provably generalize. Poster #913 this afternoon at #ICML, come hear about it!

Mohamad Amin Mohamadi@QuelMohamadAmin

What causes 𝙜𝙧𝙤𝙠𝙠𝙞𝙣𝙜 on modular addition problems? Our #ICML2024 work identifies the 𝙥𝙚𝙧𝙢𝙪𝙩𝙖𝙩𝙞𝙤𝙣 𝙚𝙦𝙪𝙞𝙫𝙖𝙧𝙞𝙖𝙣𝙘𝙚 of the task as the root cause of poor generalization early in training. Paper: arxiv.org/abs/2407.12332

English

0

1

12

2.6K

Danica Sutherland รีทวีตแล้ว

Gautam Kamath@thegautamkamath·25 Haz

Previous live link is now private. Cut version is here: youtu.be/O2gpl5l2eQA?si…

YouTube

English

0

1

9

1.5K

Danica Sutherland รีทวีตแล้ว

Arthur Gretton@ArthurGretton·21 Şub

Testing for conditional independence is notoriously difficult. Here's o̶n̶e̶ three weird tricks to: 👾 make testing work better on limited data, 👾 get correct false positive rates arxiv.org/abs/2402.13196 @rmnpogodin,@AntoninSchrab,@yazhe_li,@d_j_sutherland

English

1

17

57

8.7K

Danica Sutherland รีทวีตแล้ว

MI JUNG PARK@MIJUNGPARK11·11 Ara

We're hiring a Postdoc and two Ph.D. students to work together on privacy problems in ML at DTU. (and yes, I moved again, lol) efzu.fa.em2.oraclecloud.com/hcmUI/Candidat… efzu.fa.em2.oraclecloud.com/hcmUI/Candidat…

English

0

4

16

2K

Danica Sutherland@d_j_sutherland·27 Tem

@HamedShirzad13 @ameya_pa @ijalabv Teaser thread about Exphormer :)

Ameya Velingker | अमेय वेलिंगकर@ameya_pa

1/ 🧵I'm pleased to announce Exphormer: Sparse Transformers for Graphs (with my fantastic coauthors @HamedShirzad13 @ijalabv @d_j_sutherland and Ali Sinop), which appears at ICML 2023. Come check out our poster at 10:30am HST on July 27! arxiv.org/abs/2303.06147

English

0

2

624

Danica Sutherland@d_j_sutherland·25 Tem

@QuelMohamadAmin @HamedShirzad13 @ameya_pa @ijalabv Plus, secret bonus: this morning (#704) I'll tell people about goodness-of-fit testing for sequential models (eg strings), arxiv.org/abs/2210.10741 by Jerome Baum, Heishiro Kanagawa, @ArthurGretton.

English

1

7

4.7K

Danica Sutherland@d_j_sutherland·25 Tem

Come say hi at #ICML2023! - This afternoon (#837): orders-of-magnitude speedup in evaluating empirical NTKs, with guarantees arxiv.org/abs/2206.12543 - Thursday morning (#229): Exphormers, scalable graph transformers based on expander graphs arxiv.org/abs/2303.06147

English

2

5

27

3.7K

Danica Sutherland รีทวีตแล้ว

Ameya Velingker | अमेय वेलिंगकर@ameya_pa·27 Tem

1/ 🧵I'm pleased to announce Exphormer: Sparse Transformers for Graphs (with my fantastic coauthors @HamedShirzad13 @ijalabv @d_j_sutherland and Ali Sinop), which appears at ICML 2023. Come check out our poster at 10:30am HST on July 27! arxiv.org/abs/2303.06147

English

1

3

11

2.7K

Danica Sutherland@d_j_sutherland·25 Tem

@soldni yes!

0

1

55

Luca Soldaini 🎀@soldni·25 Tem

@d_j_sutherland danicaaaaa let’s hang!!

English

1

0

1

155

Danica Sutherland@d_j_sutherland·23 Mar

If you want to avoid accidentally uploading comments in your tex source to arXiv, consider using a script like github.com/djsutherland/a… that strips them automatically (among other niceties). :p

DV@DV2559106965076

You might know that MSFT has released a 154-page paper (arxiv.org/abs/2303.12712) on #OpenAI #GPT4 , but do you know they also commented out many parts from the original version? 🧵: A thread of hidden information from their latex source code [1/n]

English

0

2

10

2K

Danica Sutherland@d_j_sutherland·25 Oca

@katiedimartin @David_desJ as of 2014 seems like some business/econ profs were close (for MIT-only salaries, not counting any outside stuff), but couldn't find more recent numbers or stuff for any other department thetech.com/2016/06/02/mit…

English

1

0

113

Danica Sutherland

ค้นพบ