Baharan Mirzasoleiman
@baharanm
Followers
2K
Following
266
Media
4
Statuses
77
Assistant professor @UCLAComSci. Better ML via better data, Machine learning, Optimization
Los Angeles, CA
Joined July 2018
We’re thrilled by the amazing response to our #ICML2024 tutorial on “Foundations of data-efficient learning”! Over 1000 attendees joined us. Thank you all! 🙌🌱🌱🌱. ➡️ Slides: ➡️ Recording: will be available on Aug 22 🎊🎊
I'll be giving a 2-hour tutorial on data-efficient learning with my PhD student @sjoshi804 on Monday July 22 at #ICML2024. Join us to learn more about this cool topic! ➡️ We can learn better from better data! ⬅️🙌🌱
6
18
216
Can weak LLMs supervise strong LLMs to obtain superior performance? 🤔 Yes!! 🤩. Which weak models are better supervisors? 🤔. Check out @xue_yihao65785’s awesome #icml2025 paper to know how to identify best weak supervisors without having to collect labels! 🎉🌱.
🎉 Our paper “Representations Shape Weak-to-Strong Generalization” is accepted at #ICML2025!.We study weak-to-strong generalization (W2SG)—a core problem in superalignment—and offer new insights into the role of models' internal representations in W2SG. 1/
1
2
25
RT @SCSLWorkshop: 🚨 Join us at the Workshop on Spurious Correlation & Shortcut Learning (SCSL) at #ICLR2025!.@iclr_conf .🗓️ April 28, 2025….
0
15
0
Can we pretrain deep models with small synthetic data? . Dataset Distillation via Knowledge Distillation is the way to go!. Check out @sjoshi804’s #ICLR2025 paper this Saturday April 26 at 9am, Poster #307 🎉🌱.
1
4
37
Want to train LLMs with better performance and lower GPU memory requirements on data mixtures?. Check out this cool #ICLR2025 paper of @dangnth97 this Friday April 25 at 10am, Poster #265.🎉🌱.
🎉 Achievement unlocked: having papers with all of my labmates and somehow all ended up at ICLR!. I’ll be presenting our work “Mini-batch Coresets for Memory-efficient Language Model Training on Data Mixtures” at #ICLR2025 🇸🇬. Come by and chat! 👋 on Fri, Apr 25 | 10 AM GMT+8.
1
0
19
Big congrats @YuYang_i on your graduation!! 🎉🎉 🎉.very nice PhD thesis with great contributions 🌱.I’m proud of all you’ve done, and I wish you the best! 💝.
Sharing a little late update (before it’s no longer news): I wrapped up my PhD at the end of last year and recently joined @OpenAI’s reasoning team 🍓✨!.
0
0
55
RT @TheGradient: (2/2) Not at UCLA but interested in this work? Check Thanks to our fantastic intern @unregularize….
arxiv.org
Curvature regularization techniques like Sharpness Aware Minimization (SAM) have shown great promise in improving generalization on vision tasks. However, we find that SAM performs poorly in...
0
1
0
RT @andrew_ilyas: At NeurIPS? Check out the 2nd workshop on Attributing Model Behavior at Scale (ATTRIB)!. Meeting Rm 205-207, starting @ 9….
0
8
0
I’ll help presenting our #NeurIPS2024 posters tomorrow (Friday):🌱. 1- Changing the training data distribution to improve in-distribution performance (11@west #7106) w. @dangnth97. 2- Data selection for fine-tuning LLMs with superior performance (16:30@west #5401) w. @YUYANG_UCLA.
2
5
59
Same training and test distribution yields optimal in-distribution performance?. @dangnth97 showed in his #NeurIPS2024 paper that this is not true when training with gradient methods!!😮🙃.Changing the training data distribution yields SOTA!🎊. Check it out Fri Dec 13, 11am, PS#5.
0
1
23
Smaller high-quality subsets of language data not only improve LLMs’ training efficiency, but also yield considerably better performance! 🙌🎉🌱. @YUYANG_UCLA has a theoretically-rigorous method for this in her #NeurIPS2024 paper!. Check it out on Fri, Dec 13, 16:30, #PS 6.
1/ I'll be at #NeurIPS2024 presenting our work SmallToLarge (S2L): Data-efficient Fine-tuning of LLMs! 🚀. What’s S2L? It’s a scalable data selection method that trains a small proxy model to guide fine-tuning for larger models, reducing costs while preserving performance. 👇
0
9
48
RT @UCLAengineering: Assist. Prof. Baharan Mirzasoleiman @baharanm of @UCLAComSci & her large-scale machine learning research group @UCLA i….
cns.utexas.edu
Stella Offner of University of Texas at Austin Astronomy along with Arya Farahi of the Department of Statistics and Data Sciences lead the new CosmicAI.
0
1
0
RT @_xiang_chen_: 📢 @UCLAComSci is hiring! Open to all CS areas!. - Multiple Tenure-track Assistant Professor Positions: .
0
27
0
RT @pinyuchenTW: The Adversarial Machine Learning Rising Star Awards deadline is in two weeks! Submit your application and help us promote….
0
2
0
I’ll also present “SafeClip” on behalf of @WenhanYang0315 tomorrow at 1:30pm (poster session 6) #814. See you there! 🙌.
CLIP is highly sensitive to data poisoning and backdoor attacks. In this #ICML2024 paper, @WenhanYang0315 proposed an interesting way to pretrain CLIP robust to such attacks without compromising the performance! 🌱🌱.🔗Thu, July 25, Poster session 6, #814
0
2
18
I’ll present “MixPro” on behalf of @xue_yihao65785 tomorrow at 11:30 (poster session 5) poster #800. Come check it out 🙌.
ML models are sensitive to distribution shift. Can we adapt a model with only a few examples from the target domain? In this #ICML2024 paper, @xue_yihao65785 proposes an effective way, with nice theoretical analysis🌱.🔗Thu, July 25, Poster session 5, #800
0
1
18
RT @AdvMLFrontiers: 📢 We're back with a new edition, this year at.@NeurIPSConf in Vancouver! . Paper deadline is August 30th, we are look….
0
7
0