Sharon Y. Li @SharonYixuanLi Twitter profile

Last Seen Profiles

@castorcillo_27

@matheuscajaiba

@ZCFoficial

@Jus_Showoff

@l9CBAjf44nGU4Yl

@slot_mintia

@masu_ba

@OfficialFFComic

@ISefati

@stwmaniax

@minster88661

@ITriippy

@ohhmatty

@hvdesmusic

@Seventics

@1JohnAchterberg

@ita7_TKM

@svdielius

@BandarStw

@Op__Nobita

@sebastiankehl

@kontensultaaan

@MushJoeythe

@ElsaEsnoultOff

@z25lora

@y3dcouple

@funya

@kirinagi35

@Riddlemethiss84

@badiassad

@PSG_Lotfi

@MKusagawa40837

@murthuj22641169

@M7madal5aldii

@DailyRerunGf

@naohironiiyama

Sharon Y. Li

@SharonYixuanLi

8 months

Many AI researchers today display signs of burn out. Companies are racing to build bigger models, individuals rush to publish more papers. I miss the old days when things were slow. less noise, less hype. More time to savor. Collectively as a community, are we happier?

73

175

2K

Sharon Y. Li

@SharonYixuanLi

4 years

My deep learning course now has lectures available online, please check out our YouTube channel: Topics covered this fall: reliable deep learning, generalization, learning with less supervision, lifelong learning, deep generative models and more.

CS 839: Special Topics in Deep Learning

www.youtube.com

12

239

1K

Sharon Y. Li

@SharonYixuanLi

4 years

Suffering from overconfident softmax scores? Time to use energy scores! Excited to release our NeurIPS paper on "Energy-based Out-of-distribution Detection", a theoretically motivated framework for OOD detection. 1/n Paper: (w/ code included)

17

267

1K

Sharon Y. Li

@SharonYixuanLi

2 years

Sharing our new #ICML2022 paper on Logit Normalization (LogitNorm), a simple fix to the cross-entropy loss that mitigates the overconfidence issue of deep neural networks. (1/n) Paper: . 5-min video:

Mitigating Neural Network Overconfidence with Logit Normalization@I...

null

www.youtube.com

5

74

533

Sharon Y. Li

@SharonYixuanLi

5 months

Can we align LLMs without retraining the model (e.g. using RLHF)? Introducing 🔥ARGS🔥, a simple and powerful test-time alignment approach that leverages a reward model to "guide" your unaligned LLM in decoding time! 🧵(1/n) #ICLR2024 Paper:

4

67

385

Sharon Y. Li

@SharonYixuanLi

2 years

How can we make neural networks learn both the knowns and unknowns? Check out our #ICLR2022 paper “VOS: Learning What You Don’t Know by Virtual Outlier Synthesis”, a general learning framework that suits both object detection and classification tasks. 1/n

7

65

369

Sharon Y. Li

@SharonYixuanLi

1 year

Honored to receive the NSF CAREER award, which will support our work on new foundations for safe and long-term beneficial learning algorithms in the open world. Thanks to all my students, collaborators, and panelists for making this job rewarding. Lessons learned below: 1/n

30

8

341

Sharon Y. Li

@SharonYixuanLi

4 years

Wrote a blog on automating the art of data augmentation, featuring latest works on the practice, theory and new direction of data augmentation from @HazyResearch . Check out on @StanfordAILab website:

4

83

319

Sharon Y. Li

@SharonYixuanLi

1 year

Vision-language models such as CLIP is powerful in zero-shot classification. But do they know what they don’t know? We investigate the promises and AI safety of large pre-trained models when it comes to out-of-distribution data. 1/n #NeurIPS2022 Paper:

3

63

262

Sharon Y. Li

@SharonYixuanLi

1 year

If you are submitting to NeurIPS, please consider avoiding title like “X is all you need”. It goes against the spirit of science, which is acknowledging and exploring the many possibilities that we don’t know yet.

7

11

247

Sharon Y. Li

@SharonYixuanLi

10 months

Grateful to be named as Innovator of the Year by MIT @techreview , for “pioneering research in the critical field of AI safety”. It’s encouraging to witness the growth of the field over the years, with now an active community contributing to the space.

2023 Innovator of the Year: As AI models are released into the wild, Sharon Li wants to ensure...

Li’s research could prevent AI models from failing catastrophically when they encounter unfamiliar scenarios.

www.technologyreview.com

26

12

239

Sharon Y. Li

@SharonYixuanLi

1 year

Embedding quality is the key to distance-based OOD detection. Here is a more elegant alternative to off-the-shelf contrastive loss. Sharing our #ICLR2023 paper “How to Exploit Hyperspherical Embeddings for Out-of-Distribution Detection?”. 1/n Arxiv:

3

45

230

Sharon Y. Li

@SharonYixuanLi

2 years

My lab has Ph.D. openings starting in Fall 2023. I am looking for curiosity-driven students who are passionate about open-world machine learning. The deadline to apply is December 15: . Please help RT/spread the word. Thanks!

Graduate Admissions

Why choose the UW-Madison Computer Sciences MS/PhD program? Graduate students at UW-Madison have the opportunity to create new knowledge by performing research with leaders in every branch of...

www.cs.wisc.edu

5

56

214

Sharon Y. Li

@SharonYixuanLi

3 years

Time to move from CIFAR benchmarks towards OOD detection in a real-world setting! Releasing our CVPR oral paper "MOS: Scaling Out-of-distribution Detection for Large Semantic Space” 1/n Paper (w/ Rui Huang): Blog:

3

26

196

Sharon Y. Li

@SharonYixuanLi

2 years

Grateful to receive an Amazon Research Award for my proposal "Uncertainty-aware Deep Learning for Reliable Decision Making in an Open World" at @WisconsinCS . Learn more about the program on the @AmazonScience website: #AmazonResearchAwards .

Fall 2021 and winter 2022 Amazon Research Awards recipients announced

Awardees represent more than 30 universities in eight countries. Recipients have access to Amazon public datasets, along with AWS AI/ML services and tools.

www.amazon.science

21

11

191

Sharon Y. Li

@SharonYixuanLi

3 years

Both of our #CVPR2021 submissions are accepted, including one oral presentation. Thanks to our anonymous reviewer for saying "the paper is enjoyable to read" - worth all that polishing effort. Excited to release the paper and code sometime soon. Congrats to the team! :)

11

0

172

Sharon Y. Li

@SharonYixuanLi

2 years

#ICLR2022 is here. Our paper PiCO has received ICLR Oustanding Paper Award (honorable mention). Congratulations to @Haobo_zju and the entire PiCO team! The oral presentation is taking place today April 25: 7:15pm-7:30pm CDT. Paper:

7

25

159

Sharon Y. Li

@SharonYixuanLi

10 months

Given the rapid changes in AI, a major update was done for CS762 (Advanced Deep Learning). We will cover transformers, safety and alignment, foundation models and emergent behaviors, distributional shifts, and diffusion models. Tentative fall schedule:

2

14

140

Sharon Y. Li

@SharonYixuanLi

2 years

Incredibly honored to receive the @AFOSR Young Investigator Program award this year. This will support our effort on open-world machine learning and AI alignment. Many thanks to the program office and everyone who made this possible! I am grateful 💕

AFOSR

@AFOSR

2 years

Congratulations to this year's AFRL/AFOSR Young Investigator Research Program (YIP) award recipients! 🎉 @AFResearchLab #BasicResearch #AFOSRBoldResearch #AFOSRYIP #EarlyCareer #Grants #Science #Engineering

2

5

29

15

4

136

Sharon Y. Li

@SharonYixuanLi

2 months

My student Yifei Ming @ming5_alvin successfully defended his PhD thesis today! His thesis "Reliable Foundation Models in the Open World" addresses critical problems we face today in deploying large pre-trained models into the real world. Here is an overview of representative

4

130

Sharon Y. Li

@SharonYixuanLi

1 year

My student @YiyouSun has successfully defended his PhD thesis today! Yiyou has made major contributions to the field of OOD detection and open-world ML, which advance and formalize our understanding in this area. Congrats Dr. Sun, for this incredible journey 🍻

9

2

131

Sharon Y. Li

@SharonYixuanLi

2 years

Peanut passed away yesterday. He created so many happy memories since he entered my life in November 2013. We witnessed each other’s growth, shared all the life transitions, from start of my PhD to work, from Ithaca to Madison. He is the sweetest friend and will be greatly missed

12

0

131

Sharon Y. Li

@SharonYixuanLi

9 months

Ever wondered what the world looks like beyond your training data? Thrilled to release @xuefeng_du 's latest #NeurIPS2023 paper: DREAM-OOD, a cool framework for crafting photo-realistic OOD images from any in-distribution dataset. Dive in! [1/n] 📄 Paper:

2

27

129

Sharon Y. Li

@SharonYixuanLi

4 years

Together with @balajiln @DanHendrycks @tdietterich and @latentjasper , we will be organizing a workshop on "Uncertainty & Robustness in Deep Learning" at #ICML2020 . See for more info. Please submit your work (deadline: May 22, 2020) and attend the workshop!

ICML UDL 2020

Overview

sites.google.com

1

22

124

Sharon Y. Li

@SharonYixuanLi

3 years

Interested in learning about new works on out-of-distribution detection? Please join us and chat at the #NeurIPS2021 poster session next week. Hope to see you there!

0

16

118

Sharon Y. Li

@SharonYixuanLi

8 months

Early days of my grad school, watching TED talks gave me so much inspiration and taught me public speaking. Tonight I am fortunate to have the opportunity to pay back and share my journey. It’s certainly a dream of my 20s coming true. Thanks #TEDxUWMadison for the great event

3

7

120

Sharon Y. Li

@SharonYixuanLi

4 years

Kicked off my first lecture at @WisconsinCS today. What an interesting and strange time to start a tenure-track. Kudos to Blackboard Collaborate which has made the online teaching experience a breeze. I still like in-person talk and dynamics better (to at least see the crowd).

3

120

Sharon Y. Li

@SharonYixuanLi

2 months

Alignment techniques are crucial for large language models like GPT, Llama, etc. Yet, the theoretical understanding of alignment is still in its infancy. @shawnim00 's recent work, accepted by #ICML2024 , takes an exciting step in this direction. Here, I reflect on our recent

Shawn Im

@shawnim00

3 months

Can we rigorously understand how models learn behaviors through preference learning (RLHF, DPO)? 🤔 We look into this question and find that the training dynamics have a way of prioritizing behaviors! Paper: [1/n]

2

15

82

0

12

111

Sharon Y. Li

@SharonYixuanLi

2 years

First time flying on the lake

7

1

108

Sharon Y. Li

@SharonYixuanLi

1 year

Made it to Honolulu for #ICML2023 ! Waiting for the sunrise in darkness while being 5 hours jet-lagged. 🌄 Really excited to meet the new and old friends on the island. Ps. We will be presenting 3 papers at the main conference. Stop by and chat or DM me for a coffee meetup?

0

4

99

Sharon Y. Li

@SharonYixuanLi

3 months

This work is led by two incredibly talented undergraduate students: @KhanovMax a sophomore at UW Madison and recently won the prestigious Goldwater Scholarship @top34051 spent junior & senior years with us and is now pursuing graduate study at @Stanford CS. He will be traveling

Sharon Y. Li

@SharonYixuanLi

5 months

Can we align LLMs without retraining the model (e.g. using RLHF)? Introducing 🔥ARGS🔥, a simple and powerful test-time alignment approach that leverages a reward model to "guide" your unaligned LLM in decoding time! 🧵(1/n) #ICLR2024 Paper:

4

67

385

2

5

97

Sharon Y. Li

@SharonYixuanLi

1 year

If you are looking for adventures before ICML starts, drive west all the way to the Ka’ena Point Trailhead (where the road ends) and hike along the shore. Keep an eye out for dolphins, we saw a dozen today. Mountains are spectacular on the way too.

3

2

93

Sharon Y. Li

@SharonYixuanLi

8 months

NSF (in partnership with Open Philanthropy and Good Ventures) is going to fund our new project on building foundations for safety-aware machine learning. Glad to see more national-level initiatives emphasizing research on safe artificial intelligence.

6

93

Sharon Y. Li

@SharonYixuanLi

3 years

Honored to receive a Facebook (now Meta) Research Award on safeguarding neural networks. I am even more grateful that the industry world recognizes the importance of the problem and actively supports our effort in building reliable AI.

Announcing the winners of the Building Tools to Enhance Transparency in Fairness and Privacy RFP |...

In August, Meta, formerly known as Facebook, launched the Building Tools to Enhance Transparency in Fairness and Privacy request for proposals (RFP). Today, we’re announcing the winners of this award.

research.facebook.com

3

4

82

Sharon Y. Li

@SharonYixuanLi

4 years

In response to the recent challenging situations, the Uncertainty & Robustness in Deep Learning (UDL) workshop submission deadline is extended to June 14. See for more information. w/ @balajiln @DanHendrycks @tdietterich @latentjasper @icmlconf . #icml2020

ICML UDL 2020

Overview

sites.google.com

2

19

80

Sharon Y. Li

@SharonYixuanLi

2 months

My students and collaborators will present 4 exciting papers on reliable ML at #ICLR2024 . If you are attending in Vienna, please check them out! 🥰 1⃣ ARGS: Alignment as Reward-Guided Search ( @KhanovMax , @top34051 ) A test-time decoding framework that integrates alignment into

1

5

79

Sharon Y. Li

@SharonYixuanLi

1 year

Proud advisor moment: @xuefeng_du has received the inaugural Jane Street Fellowship. Congratulations!

Yaron (Ron) Minsky

@yminsky

1 year

Exciting news! Jane Street has announced the winner's of its first Graduate Research Fellowship: It was a great process, and we were all deeply impressed with the quality of the applicants.

3

94

2

3

73

Sharon Y. Li

@SharonYixuanLi

2 months

Human values encompass far more than "helpfulness" and "harmlessness". They span broad personality traits, political views, moral beliefs, and beyond. Can we elicit diverse personas encoded in LLMs? Check out our #ICML2024 paper PICLe: Persona In-Context Learning 🥒 (with

Hyeong-Kyu Froilan Choi

@HyeonggyuC

2 months

Can we modify the behavior of your LLM without training? Introducing PICLe🥒 We elicit diverse personas from LLMs with just a few demonstrative examples! #icml2024 Paper: (with @SharonYixuanLi ) [1/6]

1

12

62

0

8

70

Sharon Y. Li

@SharonYixuanLi

2 years

Tomorrow I will give a talk at the Anomaly Detection for Scientific Discovery (AD4SD) Seminar. I will share some thoughts on the opportunities and challenges in out-of-distribution detection. Talk info and link available:

2

6

66

Sharon Y. Li

@SharonYixuanLi

10 months

Thanks @techreview @Melissahei for this article. Humbled to be featured with this group of great minds working on some of the most pressing problems in AI and beyond.

Meet the next generation of AI superstars

Plus: The cofounder of DeepMind on the future of AI.

www.technologyreview.com

1

4

58

Sharon Y. Li

@SharonYixuanLi

2 years

This was the first paper Yiyou and I wrote together when he joined my group in fall 2020. I’ve always remembered the excitement we had in this idea, despite a few rejections. He has done several other excellent works ever since. What a journey, looking back.

Yiyou Sun

@YiyouSun

2 years

Do we really need all those weight parameters for OOD detection? Excited to share our #ECCV2022 paper DICE – a new sparsification-based framework for OOD detection. 1/n (joint work with @SharonYixuanLi )

1

16

79

1

4

57

Sharon Y. Li

@SharonYixuanLi

3 years

Thanks to the senior faculty from @WisconsinCS for the surprise in mail! Such a thoughtfully curated collection of local gift and a warmly written letter. Made my day!

0

1

56

Sharon Y. Li

@SharonYixuanLi

1 year

Poppy is growing up fast and likes to “loaf” around just like Peanut used to. The sunroom has become the cats room.

2

1

55

Sharon Y. Li

@SharonYixuanLi

2 years

I am giving an invited talk at @eccvconf workshops today on "How to Handle Data Shifts? Challenges, Research Progress, and Path Forward". Join us at: 1. Uncertainty Quantification for CV: 2. Learning from Limited and Imperfect Data:

UNCV Workshop @ ECCV 2022

ECCV 2022 Workshop on Uncertainty Quantification for Computer Vision.

uncv2022.github.io

1

6

55

Sharon Y. Li

@SharonYixuanLi

3 years

A comprehensive survey on OOD detection and beyond. We hope this can be a useful resource for you to learn about the connections and differences among these topics, and find relevant literature in one place.

Jingkang Yang @NTU🇸🇬

@JingkangY

3 years

Outlier Detection? Anomaly Detection? Novelty Detection? Open Set Recognition? OOD Detection? 🤨 What are they?🤔 Are they different?🧐 How to solve them?😕 Check out our latest survey "Generalized OOD Detection" to answer them all!

5

47

230

0

2

55

Sharon Y. Li

@SharonYixuanLi

1 month

📢Excited to share our latest #ICML2024 paper, which bridges understanding between classical anomaly detection and modern OOD, revealing the importance of leveraging labeled ID data. Ensuring the reliability of machine learning involves detecting data points straying from the

Xuefeng Du

@xuefeng_du

1 month

Anomaly detection and OOD detection have been widely studied, but differ in the use of in-distribution (ID) labels during training. This raises a fundamental question: How and when does ID label help OOD detection? 💡 Our #ICML2024 paper provides a formal understanding on this!

1

13

49

0

6

52

Sharon Y. Li

@SharonYixuanLi

1 year

I had fun attending the MIT conference on mechanistic interpretability and AI safety. Small and focused conference is a charm (and *every* participant gets the the time to introduce themselves) My talk and full program is available at :

0

3

45

Sharon Y. Li

@SharonYixuanLi

2 years

Flying for #ICML2022 today. Let’s catch up if you are also attending in person. I will - Share a few new works on OOD detection - Give a talk at the DataPerf workshop (7/22) - Help @ml_angelopoulos and @stats_stephen organize the #DFUQ workshop (7/23) See you there!

0

1

39

Sharon Y. Li

@SharonYixuanLi

1 year

Thank you @Casmi_NU for featuring our #ACL2023 work "Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection” This is also @RUppaal 's first paper with the lab - really excited for her!

0

4

40

Sharon Y. Li

@SharonYixuanLi

1 year

As a follow-up, @ming5_alvin and I looked into this problem since last fall: "How does fine-tuning impact OOD detection in vision-language models"? Our findings are now summarized in this article:

How Does Fine-Tuning Impact Out-of-Distribution Detection for...

Recent large vision-language models such as CLIP have shown remarkable out-of-distribution (OOD) detection and generalization performance. However, their zero-shot in-distribution (ID) accuracy is...

arxiv.org

Sharon Y. Li

@SharonYixuanLi

1 year

Vision-language models such as CLIP is powerful in zero-shot classification. But do they know what they don’t know? We investigate the promises and AI safety of large pre-trained models when it comes to out-of-distribution data. 1/n #NeurIPS2022 Paper:

3

63

262

1

4

38

Sharon Y. Li

@SharonYixuanLi

2 months

Congratulations to my student @shreym0di for winning the prestigious David DeWitt Undergraduate Scholarship! This is the premier and most competitive scholarship for undergraduates in Computer Science, and it recognizes academic excellence within our department. In Shrey’s own

0

1

39

Sharon Y. Li

@SharonYixuanLi

3 years

CS Visit Weekend is happening tmr! If you are attending, I highly encourage you to interact with our faculty and students. I'd be glad to answer questions about @WisconsinCS or my research. The best way to make an informative decision is to talk to people and get perspectives :)

0

1

38

Sharon Y. Li

@SharonYixuanLi

2 years

what a surprise 🙈🙈🙈

Martin Görner

@martin_gorner

2 years

Looks like this paper is going to the stratosphere. #1 on Arxiv now. Congratulations @SharonYixuanLi , @xuefeng_du , @MuCai7 .

1

18

0

37

Sharon Y. Li

@SharonYixuanLi

2 years

Yifei and Ying will be giving their ICML oral talk this afternoon. We formalize outlier mining as a sequential decision-making problem and show Thompson sampling can effectively balance exploitation vs. exploration for selecting informative outliers.

0

4

35

Sharon Y. Li

@SharonYixuanLi

4 years

Please join us at #UDL2020 poster session and chat about our latest work on ATOM (Adversarial Training with Informative Outlier Mining). w/ @jiefengchen1 , @andrewxiwu , @YingyuLiang1 , @jhasomesh

Jiefeng Chen

@jiefengchen1

4 years

Attending #ICML2020 ? Check out our poster at Uncertainty & robustness workshop (UDL), and learn about our latest SOTA results on using informative outlier mining for out-of-distribution detection :-) The session starts at 9am PT. Hope to see you there!

1

10

1

6

33

Sharon Y. Li

@SharonYixuanLi

3 years

Please consider submitting to our UDL workshop this year! We look forward to your contribution.

Balaji Lakshminarayanan

@balajiln

3 years

Excited to announce that we'll be organizing a workshop on "Uncertainty & Robustness in Deep Learning" at @icmlconf this year! Submissions due June 2, 2021. More details: cc @DanHendrycks @SharonYixuanLi @latentjasper @tdietterich @csilviavr @sebnowozin

5

90

492

0

31

Sharon Y. Li

@SharonYixuanLi

3 years

Reconsider the plant challenge after failing a succulent. 🐱 seems concerned whether I can keep them in shape for long. Let’s see.

1

0

31

Sharon Y. Li

@SharonYixuanLi

4 years

The video recording of my talk at MLOS seminar is available online! You can learn a series of our works on the algorithm and theory of out-of-distribution detection in 1 hour: (1) ODIN (2) Energy OOD (3) ATOM (informative outlier matters, current SoTA)

MLOS Seminar (9/21): Sharon Li

Professor Sharon Li's talk on Real Open-World Data. See https://today.wisc.edu/events/view/153038 for talk details.

www.youtube.com

2

6

31

Sharon Y. Li

@SharonYixuanLi

3 years

I will be giving a talk at the NeurIPS'21 workshop on challenges and opportunities in uncovering unknowns for the ImageNet model. Exciting agenda (Dec 13) put together by the organizers @zeynepakata @coallaoh @dlarlus @giffmana @SanghyukChun @XiaohuaZhai

Lucas Beyer (bl16)

@giffmana

3 years

Our NeurIPS'21 workshop on "ImageNet: past, present, and future" has been accepted! I'm excited about our speaker line-up. I'm even more excited to see what papers researchers will submit to the workshop! Please spread the word, and consider submitting.

2

44

254

0

8

31

Sharon Y. Li

@SharonYixuanLi

2 years

Great to see the enthusiasm in our work on unknown-aware object detection using videos in the wild. Work led by my awesome student @xuefeng_du . We sadly missed #CVPR2022 in person but the oral talk is recorded online. Get in touch with us if you are interested in chatting.

Xuefeng Du

@xuefeng_du

2 years

Excited to release our #CVPR2022 oral paper STUD, a powerful unknown-aware object detection framework that safeguards against OOD objects. STUD is the first to leverage videos in the wild and teaches models to tell apart known and unknowns. (1/n) Paper: .

3

14

91

0

2

30

Sharon Y. Li

@SharonYixuanLi

3 years

Thank you @amfam and @datascience_uw for sponsoring our research efforts on building #ResponsibleAI . Both estimating distributional uncertainty and debiasing ML are timely problems to tackle.

datascience@uw

@datascience_uw

3 years

. @amfam has partnered with @UWMadison through the American Family Insurance Data Science Institute to offer mini-grants for data science research. Nearly $3 million has been awarded to 21 teams since 2020. Learn about Round 3 awards:

0

3

16

1

0

29

Sharon Y. Li

@SharonYixuanLi

6 months

Excited that our work is now available at Nature Scientific Reports!

Kaiping Chen

@Kaiping_Chen

6 months

Thrilled to share our latest work in @SciReports on conversational AI. We assessed how #GPT interacts with diverse social groups on science & social issues, introduced an equity framework and shared our datasets. Full study is here: #scicomm #OpenAI #HCI

3

26

73

0

2

27

Sharon Y. Li

@SharonYixuanLi

3 years

Look forward to speaking at the seminar! Thank you for hosting.

Gary Shiu

@gary_shiu

3 years

We are pleased to have our very own Sharon Li @SharonYixuanLi @WisconsinCS as our speaker of the seminar this week. She will tell us about "Uncovering the Unknowns of Deep Neural Networks: Challenges and Opportunities" on Dec 1 at noon ET. @UWMadPhysics

0

4

19

1

3

28

Sharon Y. Li

@SharonYixuanLi

4 years

I took my first lecture at Cornell with this wise man, who later became my phd advisor. He would always be in his office by 9am (even if he was on an international flight the day before). He stands for what a dedicated and disciplined career truly means.

Jeff Dean (@🏡)

@JeffDean

4 years

Congrats on the last lecture in a long and distinguished career to John Hopcroft! I've gotten to spend a bit of time w/John at @HLForum over last few years, & he's a delight to be around. (Sorry the last lecture wasn't in a classroom filled w/students to send you off properly!)

3

13

133

1

0

27

Sharon Y. Li

@SharonYixuanLi

8 months

@ShuiwangJi Yupp. And perhaps less is more.

1

0

26

Sharon Y. Li

@SharonYixuanLi

3 years

Panel taking place now w/ Michael I. Jordan, Vladimir Vovk, and Larry Wasserman

Stephen Bates

@stats_stephen

3 years

🚨The Distribution-free Uncertainty Quantification ICML workshop kicks off tomorrow!🚨 Leading off the morning session will be Rina Barber, Michael Jordan, Vladimir Vovk, Larry Wasserman, and Leying Guan.

3

19

78

0

4

26

Sharon Y. Li

@SharonYixuanLi

4 years

(3/) In contrast, we show mathematically that softmax confidence score is a biased scoring function that is not aligned with the density of the inputs and hence is not suitable for OOD detection.

1

0

24

Sharon Y. Li

@SharonYixuanLi

4 years

In our Thanksgiving lab social, @berylSreya inspired me to give shoutouts to every student, for all the hard work, enthusiasm, and inspiration they bring us. The messages are the tokens of appreciation for how grateful I am to work with these awesome students at @WisconsinCS .

0

24

Sharon Y. Li

@SharonYixuanLi

4 years

Very excited to share this work on Model Patching—an end-to-end framework for improving robustness against subgroup differences, with benefits on a real-world skin cancer classification task. Thanks to my amazing collaborators @krandiash , Albert Gu and @HazyResearch !

Karan Goel

@krandiash

4 years

Preprint alert! "Model Patching: Closing the Subgroup Performance Gap with Data Augmentation" is now on arXiv! 📑Paper: 🧑‍💻Code: 📹Video: ✍️Blog: Read on to learn more (1/9)

3

37

154

0

2

24

Sharon Y. Li

@SharonYixuanLi

4 years

I will be giving a talk at Women in Computer Science (WiSC) at Stanford next Wednesday. I will talk about research on open-world machine learning and will stick around for a casual Q&A at the end. Look forward to this and thanks @StanfordWiCS for organizing!

0

3

23

Sharon Y. Li

@SharonYixuanLi

9 months

ICLR used to be a community of a few hundred people (as I was reminded by this old fb group). Almost 10 years later, 5k or even more submissions. 🫣 Good luck to those of you who are working on the deadline!

0

22

Sharon Y. Li

@SharonYixuanLi

3 years

If you are on the job market this year, consider applying! The living quality is a real charm (having spent years in both east and west coast before I moved here).

Somesh Jha

@jhasomesh

3 years

Our department @WisconsinCS is looking for faculty at all levels. Reach out if you need more information. FYI, Madison was rated as #1 city livable city. Just saying:-)

0

24

73

0

22

Sharon Y. Li

@SharonYixuanLi

2 years

Julia is giving an excellent talk at #ICML2022 on “training OOD detectors in their natural habitats”, a new framework that leverages wild data for practical OOD detection. Joint work with Julian Katz-Samuels, Julia Nakhleh and @rdnowak . Full paper:

1

21

Sharon Y. Li

@SharonYixuanLi

4 years

This is precisely how I spent my 30s birthday. Finished all episodes yesterday in the midst of CVPR deadline.

Demis Hassabis

@demishassabis

4 years

Highly recommend binge watching the new Netflix series Queen's Gambit, beautifully filmed, stunningly acted, and a pretty accurate portrayal of what it is like to be a child chess prodigy. Not surprising given that the brilliant @Kasparov63 was a consultant on it! #QueensGambit

75

226

3K

0

21

Sharon Y. Li

@SharonYixuanLi

3 years

If you are at @CVPR , please join us at the poster session (paper 6254 & 6415)! Thanks to our audience for the great questions and conversations. The morning session gave me so much to think about. #CVPR2021

Sharon Y. Li

@SharonYixuanLi

3 years

Time to move from CIFAR benchmarks towards OOD detection in a real-world setting! Releasing our CVPR oral paper "MOS: Scaling Out-of-distribution Detection for Large Semantic Space” 1/n Paper (w/ Rui Huang): Blog:

3

26

196

0

21

Sharon Y. Li

@SharonYixuanLi

3 years

Join us tonight at the WiML Un-Workshop "Does your model know what it doesn’t know? Uncertainty estimation and OOD detection in DL". @polkirichenko @AkramiHaleh and @jessierenjie will lead with an excellent tutorial talk, followed by breakout sessions. Pop in and chat together?

Jie Jessie Ren

@jessierenjie

3 years

Join us July 21 7:25 pm ET at WiML Un-Workshop at @icmlconf for a breakout session on "Does your model know what it doesn’t know? Uncertainty estimation and OOD detection in DL" Together with @polkirichenko @AkramiHaleh @sharonyixuanli @sergulaydore

0

4

17

0

2

20

Sharon Y. Li

@SharonYixuanLi

4 years

(2/) Joint work w/ Weitang Liu, Xiaoyun Wang, and John Owens. We show that energy is desirable for OOD detection since it is provably aligned with the probability density of the input—samples with higher energies can be interpreted as data with a lower likelihood of occurrence.

1

0

19

Sharon Y. Li

@SharonYixuanLi

3 years

Check out FDIT - a cool and well-executed idea that brings classic signal processing techniques to modern image translation. Fourier space can do wonders.

Mu Cai @ CVPR

@MuCai7

3 years

Excited to release our new ICCV paper on FDIT, a powerful frequency-domain image translation framework. FDIT substantially improves identity-preserving generation, producing photo-realistic high-res images. (1/n) Paper:

2

0

10

0

19

Sharon Y. Li

@SharonYixuanLi

4 years

Full syllabus is available at:

0

4

20

Sharon Y. Li

@SharonYixuanLi

2 years

100% agree. be out-of-distribution

Lex Fridman

@lexfridman

2 years

Weird people make life extra fun. Be weird. Fitting in with the crowd is overrated.

2K

4K

31K

0

19

Sharon Y. Li

@SharonYixuanLi

2 years

(9/) LogitNorm can be easily adopted in practice. It is straightforward to implement with existing deep learning frameworks, and does not require sophisticated changes to the loss or training scheme. Code and data are publicly available at .

GitHub - hongxin001/logitnorm_ood: Official code for ICML 2022: Mitigating Neural Network Overcon...

Official code for ICML 2022: Mitigating Neural Network Overconfidence with Logit Normalization - hongxin001/logitnorm_ood

github.com

1

17

Sharon Y. Li

@SharonYixuanLi

2 years

Students in my deep learning class generated this image using DALL·E 2, with the prompt "Happy Thanksgiving, students at University of Wisconsin - Madison". Pretty cool with the State Capitol in the background. (slide credit: @YepengJ , Sijia Fang and Jiahao Fan). happy holidays!

0

1

16

Sharon Y. Li

@SharonYixuanLi

4 years

This also includes a special talk by @tydsh (Facebook AI Research) on understanding deep neural networks in a teacher-student setting.

1

3

16

Sharon Y. Li

@SharonYixuanLi

4 years

Thank you for featuring, and especially SAIL blog team for helping with the editorial process. @andrey_kurenkov @siddkaramcheti

Stanford AI Lab

@StanfordAILab

4 years

Data augmentation is crucial for machine learning, so how do we do it in a principled way? Check out our latest blog post courtesy of @SharonYixuanLi and @HazyResearch new algorithms for automating the search process of augmentation techniques.

0

2

14

0

15

Sharon Y. Li

@SharonYixuanLi

2 years

My in-laws are brave people. They came to visit and went on a tandem flight adventure. What does the launching remind you of? Mario kart fans?

1

0

15

Sharon Y. Li

@SharonYixuanLi

2 years

I will be giving a talk at the National Institute of Standards and Technology on December 8. The open colloquia series discuss issues to help advance the state-of-the-art in AI measurement and evaluation. Schedule below:

AI Metrology Colloquia Series

As a follow-on to the National Academies of Science, Engineering, and Medicine workshop on Assessing and Improving AI Trustworthiness (

www.nist.gov

0

13

Sharon Y. Li

@SharonYixuanLi

4 years

#fallcolors are coming together in Madison. 🍂

0

14

Sharon Y. Li

@SharonYixuanLi

4 years

I had an awesome time serving as the #WiML2020 mentor this year. Thanks to my co-mentor @djhsu , Table 34 participants, and organizers of @WiMLworkshop for making the NeuIPS experience memorable.

WiML

@WiMLworkshop

4 years

We would like to express our sincere gratitude to the inspiring cohort of #WiML2020 MENTORS! We have over 120 mentors from academia & industry! Mentorship roundtables are divided into 3 areas Research (Tables 1–28), Career & Life Advice (Tables 29–50) and Sponsor (Tables 51–63).

3

6

95

0

14

Sharon Y. Li

@SharonYixuanLi

4 years

Look forward to giving a talk at the MLOS seminar next Monday. Check out the full agenda: . Thanks for organizing the great series @RemziArpaciD and Microsoft @krlis1337 @GraySystemsLab @MSFTResearch .

Carlo Curino

@carlo_curino

4 years

@RemziArpaciD @WisconsinCS @GraySystemsLab @SQLServer @MSFTResearch Next up we will have: @SharonYixuanLi Assistant Prof. in the CS dept of @UWMadison , talking about: “Reliable Open-World Learning Against Out-of-distribution Data" (Monday 9/21 at 2pm) 6/n

1

2

5

1

14

Sharon Y. Li

@SharonYixuanLi

4 years

Look forward to your participation in #UDL2020 this Friday! We are collecting questions for our panel discussion (11:30am-12:30pm PT), please submit here: .

Balaji Lakshminarayanan

@balajiln

4 years

Attending #ICML2020 ? Join us for our workshop on "Uncertainty and Robustness in Deep Learning" @icmlconf this Friday (July 17), co-organized w/ @SharonYixuanLi , @DanHendrycks , @latentjasper , @tdietterich . We had 140 (!) accepted papers 1/3 (contd.)

4

23

135

0

14

Sharon Y. Li

@SharonYixuanLi

4 years

(4/) Importantly, energy score can be derived from a purely discriminative classification model without relying on a density estimator explicitly, and therefore circumvents the difficult optimization process in training generative-based models such as JEM.

1

0

13

Sharon Y. Li

@SharonYixuanLi

2 years

Summer colors are stunning

1

0

14

Sharon Y. Li

@SharonYixuanLi

3 years

Received a package of swags from @HLForum Germany. Didn’t expect to reunion with my advisor this way (nice sticker note cover though). Thanks Laureate forum for the thoughtful gift!

0

13

Sharon Y. Li

@SharonYixuanLi

4 years

(8/) More broadly, our work builds on the insights and principles from energy-based models by @ylecun et al. We were also inspired by the early work on energy-based GAN by Jake Zhao, and JEM by Grathwohl et al.

1

0

13

Sharon Y. Li

@SharonYixuanLi

4 years

(6/) Results highlight: on WideResNet, the energy score reduces the average FPR95 by 18.03% on CIFAR-10 compared to using the softmax confidence score. With energy-based training, our method outperforms existing SoTA.

1

0

12

Sharon Y. Li

@SharonYixuanLi

1 year

Kudos to @xuefeng_du for being featured in @WisconsinCS news!

0

13

Sharon Y. Li

@SharonYixuanLi

4 years

(5/) Within our framework, we demonstrate that energy can be flexibly used as a scoring function for any pre-trained neural classifier as well as a trainable cost function to shape the energy surface explicitly for OOD detection.

1

0

12

Sharon Y. Li

@SharonYixuanLi

5 months

A really cool work led by @XuhuiHuangChem ’s group on leveraging OOD detection for scientific discovery and understanding protein dynamics.

Xuhui Huang

@XuhuiHuangChem

6 months

Excited to introduce TS-DART, that can automatically identify all Transition States(TS) across multiple free energy barriers in protein dynamics! Inspired by Trustworthy AI, TS-DART detects TS as Out-of-Distribution data in the hyperspherical latent space!

3

12

76

0

1

12

Sharon Y. Li

@SharonYixuanLi

5 years

I am so incredibly honored and thrilled to be on the #ForbesUnder30 Science list. Thank you to @Forbes , @TheAlexKnapp , @SquigglyVolcano and everyone along the way.

Sharon Li

Sharon Li on the 2020 30 Under 30 - Science - For machine learning to work, large amounts of data have to be labeled by people to �teach� computers what

www.forbes.com

0

11

Sharon Y. Li

@SharonYixuanLi

4 years

(7/) Previous approaches such as ODIN and Mahalanobis may have hyperparameters to be tuned. In contrast, the energy score is a parameter-free measure, which is easy to use and implement, and in many cases, achieves comparable or even better performance.

1

11