Research Scientist @Meta Applied AI, ex-CAS/CDS. Human-in-the-loop optimization, Bayesian methods, and sample-efficient learning from feedback. Stanford CS PhD.

Views my own.
Joined January 2013
Zhiyuan Jerry Lin retweeted
Check out our work on LILO led by @kasia_kobalczyk and @ItsMrLin. LILO allows users to perform black box optimization with natural language feedback. Hope to see you at the poster at #ICML2026 Paper: arxiv.org/pdf/2510.17671
Preference optimization often reduces feedback to comparisons or rankings But decision makers often know more than “A >> B” They can explicitly state their preferences and priors in words LILO is a new framework turning such feedback into signal for Bayesian Optimization 🧵
2
9
688
Zhiyuan Jerry Lin retweeted
My kindest sympathies for the Bayesian Optimization folks
Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.
11
40
1
591
122,484
Zhiyuan Jerry Lin retweeted
2024 PhD Internship opportunity! Join me and my amazing colleagues on Meta's Adaptive Experimentation team to work on Bayesian optimization, probabilistic modeling / Gaussian Processes, and sample-efficient decision making: metacareers.com/jobs/9056341…
1
28
2
125
34,666
Zhiyuan Jerry Lin retweeted
You knew that was coming: Code LLama !!! - Llama-2 tuned for code generation, debugging, etc - base models, Python-specific, and instruction-tuned. - 7B, 13B and 33B params models - Available with the same license as Llama-2 - Blog: ai.meta.com/blog/code-llama-… - Paper: ai.meta.com/research/publica… - Code: github.com/facebookresearch/… - Models: ai.meta.com/resources/models…
Replying to @syhw
We're releasing base models, Python-specialized models, and Instruct(ion following) models, all in sizes 7B, 13B, 34B params. Get the code and weights: github.com/facebookresearch/… Read the research paper: ai.meta.com/research/publica… Read the blog post: ai.meta.com/blog/code-llama-…
39
342
35
1,683
341,340
Zhiyuan Jerry Lin retweeted
📢 Latest @Nature @HSScomms publication from our team on the heterogeneous effects of social support on the adoption of FB vaccine profile frames.
1
2
19
2,204
Zhiyuan Jerry Lin retweeted
Strongest possible advice to early grad students: you need outside hobbies or projects that are meaningful to you. It is dangerous for your only source of value, identity, and friendship to be academia. You absolutely have to find something else to invest in, it’s just not safe
29
401
130
2,650
406,591
Zhiyuan Jerry Lin retweeted
(please retweet) I am taking CS PhD students! If you are interested in ML, data science, health, or inequality, please apply to Cornell :) We have an island campus in New York City and a fantastic group of students I feel lucky to collaborate with every day. Join us!
15
170
4
596
Zhiyuan Jerry Lin retweeted
Apple flexing so hard about how they gave back ports that they took away is just poetry.
402
4,801
499
57,549
Zhiyuan Jerry Lin retweeted
1
14
5
156
Zhiyuan Jerry Lin retweeted
Important additional context: - Many new survivors have come forward - CSE is considering tenuring Lasecki right now - The survivors who spoke to The Daily are some of the bravest people I know
1
19
108
Zhiyuan Jerry Lin retweeted
[1/6] I'm excited to share our new @AIESConf paper "Surveilling Surveillance: Estimating the Prevalence of Surveillance Cameras with Street View Data" w/ @KenielYao and @5harad arxiv.org/pdf/2105.01764.pdf 🧵
2
6
26
Zhiyuan Jerry Lin retweeted
Check out recent work with @KenielYao, @jgaeb1, @KonitzerTobias, @DavMicRot, @mieuque and @5harad, on the effect of polling place changes on voting behavior. Link, arxiv.org/pdf/2105.01728.pdf, and 🧵!
2
8
1
15
Zhiyuan Jerry Lin retweeted
every single park hang in San Francisco
664
3,819
1,997
37,636
Zhiyuan Jerry Lin retweeted
Hello! I'm recruiting students interested in ML/data science, especially as applied to healthcare or inequality. If this is you, please consider applying to 1) MSR internship (careers.microsoft.com/us/en/… - for current PhD students) and 2) Cornell CS PhD. Please RT.
20
234
6
621
Zhiyuan Jerry Lin retweeted
Last year at @CompPolicyLab, we created the country’s first race-blind charging platform, intended to reduce racial bias in prosecutor’s charging decisions. We’ve analyzed the data! Here’s what happened when we deployed it in a major district attorney’s office (thread)
6
33
15
100
Zhiyuan Jerry Lin retweeted
My Behavioral Data Science Group has postdoc positions available, especially around AI for health and mobile health, and deep learning for graphs, code, or behavioral time series. Come work with us! bit.ly/bdatapostdoc Please retweet and share :)
3
57
1
101
Zhiyuan Jerry Lin retweeted
Do you think you would you have discovered a gorilla hiding in plain sight in your data? We found that you are 5 times less likely to do so if you have a specific hypothesis in mind. *A hypothesis is a liability* genomebiology.biomedcentral.…
43
351
125
1,037
We're hiring! Come join our team of software engineers and data scientists to combat discrimination, reduce incarceration, and improve public policy. Read more at our job posting link here: policylab.stanford.edu/jobs/
9
14
Zhiyuan Jerry Lin retweeted
I spent a humiliating amount of time learning how to make animated graphs, just to illustrate a fairly obvious point. “Forecasting s-curves is hard” My views on why carefully following daily figures is unlikely to provide insight. constancecrozier.com/2020/04…
256
3,001
568
10,585