Mehak Dhaliwal @mehakdh1
CS PhD student @ UCSB Interested in AI for Healthcare, NLP, Fairness, and Robustness mehak126.github.io Joined December 2021-
Tweets18
-
Followers37
-
Following121
-
Likes22
This is exciting! Two independent groups have now confirmed the presence of self-bias in various tasks performed by large language models. Our paper, "Perils of Self-Feedback" (arxiv.org/abs/2402.11436), revealed that self-refinement processes may actually intensify the self-bias.
This is exciting! Two independent groups have now confirmed the presence of self-bias in various tasks performed by large language models. Our paper, "Perils of Self-Feedback" (arxiv.org/abs/2402.11436), revealed that self-refinement processes may actually intensify the self-bias.
Thank you for the interview @matteo_wong @TheAtlantic , great to hear diverse opinions about the topic @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1
Thank you for the interview @matteo_wong @TheAtlantic , great to hear diverse opinions about the topic @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1
A wonderful piece which draws on researcher perspectives and communities working on this problem. @davlanade @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @HolyLovenia @mehakdh1 + a spotlight on global multi-year efforts: @CohereForAI w Aya + @MasakhaneNLP
What do you do when someone questions your work while you're giving a talk? You do a collaboration! Coming to NAACL 24: a paper investigating when contested translations matter for text-to-image multilinguality assessment w/ @FPSLuozi from ASU arxiv.org/abs/2403.11092 1/8
🚀 Exciting News! 🚀 We had our first Real AI group meeting last week, and it was a fantastic start to our journey together! It was amazing to see everyone's enthusiasm and passion for AI. At Real AI, we seek to pioneer the next frontiers of artificial intelligence for science.
🎉 Thrilled to share my first first-author paper on adversarial transfer learning got accepted at @CVPR ! Feel free to check it out: arxiv.org/abs/2312.05716 Brief Summary ⏬
{UCSB|AI2|UW|Stanford|MIT|UofT|Vector|Contextual AI} present a survey on🔎Data Selection for LLMs🔍 Training data is a closely guarded secret in industry🤫with this work we narrow the knowledge gap, advocating for open, responsible, collaborative progress arxiv.org/abs/2402.16827
In case anyone else is digging through their bookmarks looking for these today, here are the rebuttal guides ;) docs.google.com/document/d/1cd…
3. MT detection and multi-way parallelism are promising data filtering techniques for improving multilingual data quality. This is also an important issue for AI researchers and internet content consumers as machine-generated content is ever-increasing.
3. MT detection and multi-way parallelism are promising data filtering techniques for improving multilingual data quality. This is also an important issue for AI researchers and internet content consumers as machine-generated content is ever-increasing.
2. The highly multi-way parallel translations are likely machine translations from English to lower-resource languages with significantly lower translation and content quality
2. The highly multi-way parallel translations are likely machine translations from English to lower-resource languages with significantly lower translation and content quality
Thank you @julesaroscoe @VICENews for covering our work! Read at arxiv.org/abs/2401.05749 Key takeaways- 1. A large portion of the internet is translated to many languages (multi-way parallel), with low-resource languages having significantly more translations
Thank you @julesaroscoe @VICENews for covering our work! Read at arxiv.org/abs/2401.05749 Key takeaways- 1. A large portion of the internet is translated to many languages (multi-way parallel), with low-resource languages having significantly more translations
Heads up for anyone looking to post an ACL-format preprint to arXiv on Jan 15: that's a US public holiday (MLK day), so there will be no postings that day. If you want to be able to link to your preprint before AOE Jan 15, your arXiv submission deadline is actually THIS Friday!
RAG for LLMs Great to see an overview of all the retrieval augmented generation (RAG) research that has been happening. Should be a great read for the end of the year. arxiv.org/abs/2312.10997…
Join for research in responsible and ethical AI, led by the best advisors!
Interested in AI for Science? Our team @ai_ucsb is recruiting! @HaewonJeong00 , @YaoQin_UCSB & I are hiring postdocs & PhD students & visiting researchers 😃 👉Apply to @ucsbece at ece.ucsb.edu/grad/apply before Dec 15! Yao & I are also #NeurIPS2023 : feel free to reach out!
Interested in AI for Science? Our team @ai_ucsb is recruiting! @HaewonJeong00 , @YaoQin_UCSB & I are hiring postdocs & PhD students & visiting researchers 😃 👉Apply to @ucsbece at ece.ucsb.edu/grad/apply before Dec 15! Yao & I are also #NeurIPS2023 : feel free to reach out! https://t.co/3ICVGUXamz
In particular, we are working on AI Safety (robustness & fairness), AI for Science (quantum and neuroscience), AI for Healthcare (Diabetes & Women Brain Health). Feel free to reach out if you are interested in one of these topics!
The REAL AI team @ UCSB @ai_ucsb is hiring! We have openings for PhD students/Postdoc/visiting researchers! Both Nina @ninamiolane and I @YaoQinUCSD will be at NeurIPS this week. Come to talk to us if you have interests in doing REAL AI research 😊
Conventional #recsys wisdom: "better to go wide than deep". Our paper: go infinitely-wide, compute the solution in closed-form with a single hyper-parameter, and considerably beat all SoTA. Furthermore, can you get the same performance with just 500 fake users? Yes! A thread 🧵
IreneJob @hw5vnII7n97WmH
12 Followers 273 FollowingXianjun Yang @Qnolan4
348 Followers 987 Following Phd @ucsbcs, Bachelor @tsinghua_uni, AI Safety, nlp, GenAI. Open to collaboration. Incoming Interns @Amazon and @AIatMeta (Looking for full-time jobs)Weixi Feng @weixi_feng
397 Followers 292 Following CS Ph.D. candidate @UCSB @UCSBNLP. Ex-research intern @Adobe, @Amazon. #Multimodality #ComputerVision #NLProc.Abel Aboh🇳🇬🇬.. @abel_DMChampion
640 Followers 571 Following DM @BankofEngland | Advisor @TDataKirk | Board @DataLabScotland, @WestLoCollege | SG @BigData_LDN | CM @LawScot | Ex @easyjet, @BAESystemsplc ~ My viewsRobert Scoble @Scobleizer
504K Followers 68K Following Follow me on my new podcast with AI startups, Unaligned. Tech industry color commentator since 1993. Author/Blogger. Former strategist @Microsoft.Mohamed Ahmed @m_0_a
321 Followers 880 Following Researcher at Microsoft Africa - working on task alignment and evaluation of LLMs | Ex- @benevolent_ai | Ex- NEC Labs Europe | Ex-@UCL | @so_innovateNishaanth Kanna @nishaanthkanna
36 Followers 594 Following skate to where the puck is going to be, not where it has been.Louisa Cornelis @louisacornelis
44 Followers 37 Following physics phd student at ucsb @geometric_intel, prev @onepeloton, cs and physics and art @scrippscollege @harveymuddAlfonso Amayuelas @AlfonAmayuelas
495 Followers 530 Following CS PhD Student at @ucsbNLP @ucsantabarbara 😎🌊🏄🏻♂️Zekun Li @ZekunLi0323
760 Followers 587 Following CS Ph.D. student @UCSB, ex-intern at #MSFTResearch, Meta #RealityLabs, interested in #NLProc, #LLM, and #AI4HealthYoungseok Yoon @YoungseokY87156
5 Followers 7 FollowingDeepak Nathani @deepaknathani11
410 Followers 804 Following PhD Student @UCSBNLP | Mentor @CAREMentorship | Previously: Intern @AWS AI, Pre-Doctoral Researcher @GoogleAI India, Undergrad @IITHyderabadTian Qiu @QiuTian_QT
17 Followers 82 FollowingReal AI @ UCSB @ai_ucsb
144 Followers 115 FollowingMadhav Ranka @mady619
177 Followers 160 Following Building @FincentHQ | Ex Engineering - @PayUindia | IIT Delhi '19 | Travel, Tech and LifeAlon Albalak @AlbalakAlon
888 Followers 465 Following CS PhD candidate at @ucsbNLP. Research: Data-centric AI, Efficiency in ML, NLP.Safiya Umoja Noble @safiyanoble
66K Followers 17K Following 2021 #MacFellow 🥰 | Author, Algorithms of Oppression, @[email protected] & @safiyanoble.bsky.socialWenda Xu @WendaXu2
679 Followers 319 Following PHD candidate at UCSB’s NLP lab coadvised by William Wang and Lei LiSrinath Tankasala @srinath_tank
142 Followers 479 Following Applied Scientist II at Amazon| PhD candidate in Robotics at UT Austin| opinions are my ownMichael Saxon @m2saxon
2K Followers 1K Following CS PhD cand @ucsbNLP 🌊🌴 @NSF GRFP 🧐analyzing semantics in generative lang/img AI models🤖 Big tech ex-intern. BS/MS @ASU 🌵🏜 🔜 @AMD opensrc GenAI RS internYao Qin @YaoQin_UCSB
3K Followers 272 Following Assistant Professor @UCSB; Research Scientist @googleresearch; PhD @UCSDAndong Hua @andong_1997
36 Followers 41 Following Research Assistant @UCSB; Former Research Engineer at TuSimple; Master @UCLAHesam Asadollahzadeh @HesamAsdz
82 Followers 495 Following Trustworthy & Reliable AI/ML Researcher @MLL_SharifAI & @sangerinstitute; CSE BSc Student @UnivOfTehran; Header created by DALL·E 3 :)Nina Miolane @ninamiolane
4K Followers 2K Following Assis. Prof. @ucsbece Affiliate @SLAClab Stanford Prev @Stanford @Inria @imperialcollege @Polytechnique PI @geometric_intel https://t.co/Eeeaqytppe, PilotNavreet Kaur @navreeetkaur
173 Followers 557 Following Research in NLP, Responsible AI, HCI, AI Ethics. She/herDaniel Beaglehole @dbeagleholeCS
293 Followers 347 Following ML PhD @UCSanDiego. One foot in theory and one foot in practice… in theory.Ribhav Gaur @GaurRibhav
23 Followers 589 Following Student @MIT | @IITDelhi alum | Tennis enthusiast | @rogerfederer, @DjokerNole, @RafaelNadal fanMahdi Hajiaghayi @mhajiaghayi
92 Followers 178 Following ML manqge at microsoft, focusing on Image search and recommendation, Vision and Language modelYashar Deldjoo @yashardel
765 Followers 1K Following Tenure Track (Rtd-b), Asst. Professor @polibaofficial; #RecSys #GenerativeAI #FairML #TrustworthyML #Multimedia #Fashionlovish @louvishh
315 Followers 603 Following phding @ucl and @aiatmeta (llama team). mostly random tweets here.Xianjun Yang @Qnolan4
348 Followers 987 Following Phd @ucsbcs, Bachelor @tsinghua_uni, AI Safety, nlp, GenAI. Open to collaboration. Incoming Interns @Amazon and @AIatMeta (Looking for full-time jobs)Zichen Chen (🐱,�.. @my_cat_can_code
283 Followers 179 Following CS PhD student at UCSB 🎓 | Trustworthy AI/FL/HAI 📖 | DAOlivia co-founder @dao_oliviaLouisa Cornelis @louisacornelis
44 Followers 37 Following physics phd student at ucsb @geometric_intel, prev @onepeloton, cs and physics and art @scrippscollege @harveymuddZachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷Zhuosheng Zhang @zhangzhuosheng
1K Followers 541 Following Tenure-Track Assistant Professor at @sjtu1896. NLP/AI/ML. Prev intern @AmazonScience @MSFTResearch @NICT_Publicity @sinovationvc @IBM #NLProcNick Frosst @nickfrosst
13K Followers 847 Following cofounder @cohere - singer @goodkidband pfp: @polarfishh_26UCSB NLP Group @ucsbNLP
1K Followers 729 Following The NLP Group @ University of California, Santa Barbara. Profs. @WilliamWangNLP, Xifeng Yan, Simon Todd, @CodeTerminator, @lileics; acct run by @m2saxonXinyi Wang @XinyiWang98
795 Followers 299 Following UC Santa Barbara CS PhD student working on ML/NLPWeixi Feng @weixi_feng
397 Followers 292 Following CS Ph.D. candidate @UCSB @UCSBNLP. Ex-research intern @Adobe, @Amazon. #Multimodality #ComputerVision #NLProc.Cohere For AI @CohereForAI
15K Followers 177 Following We are a research lab and open science initiative that seeks to solve complex machine learning problems. Join us in exploring the unknown, together.Masakhane @MasakhaneNLP
5K Followers 446 Following Ubuntu Punk 🌍 The community theatre of ML ✨ Putting African languages on the #NLP map since 2019Dana Lewis @Danaslewis
17K Followers 3K Following Award winning TV Correspondent and host. Creator of the hugely influential Back Story with Dana Lewis Podcast. https://t.co/xU8ab7EcJFMatteo Wong @matteo_wong
3K Followers 1K Following Associate Editor for Science, Tech, and Health @theatlanticAbel Aboh🇳🇬🇬.. @abel_DMChampion
640 Followers 571 Following DM @BankofEngland | Advisor @TDataKirk | Board @DataLabScotland, @WestLoCollege | SG @BigData_LDN | CM @LawScot | Ex @easyjet, @BAESystemsplc ~ My viewsDavid Ifeoluwa Adelan.. @davlanade
2K Followers 1K Following @DeepMind Academic Fellow @uclcs, incoming assistant Professor @mcgillu, Canada CIFAR AI Chair @CIFAR_News | interested in multilingual NLP | Disciple of JesusSara Hooker @sarahookr
39K Followers 7K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.Muhammad AbdulMageed @mageed
2K Followers 2K Following Canada Research Chair in Natural Language Processing and Machine Learning; Director of UBC Deep Learning & NLP Group. #LLMs🌴Muhao Chen🌴 @muhao_chen
2K Followers 552 Following 🐹Assistant Professor of Computer Science @UCDavis🐹 | 💙PhD @UCLAComSci 2019💛 | 🌴加州boy🌴 | 🎸@GALNERYUSOFFIC2#1🎧! |♛Collecting⌚♛Wenbo Guo @WenboGuo4
774 Followers 535 Following AP@UCSB; deep learning, security; Ph.D. from Penn State, 2020 IBM Ph.D. fellowship Awardee.Conference on Languag.. @COLM_conf
2K Followers 6 Following https://t.co/GhGCMEoa4A Abstract submission: March 22, 2024Tidepool @Tidepool_org
8K Followers 370 Following Tidepool is an open source, nonprofit effort to liberate diabetes data, support researchers, and provide great, free software to the diabetes community.Alfonso Amayuelas @AlfonAmayuelas
495 Followers 530 Following CS PhD Student at @ucsbNLP @ucsantabarbara 😎🌊🏄🏻♂️Jie Huang @jefffhj
4K Followers 569 Following Ph.D. Candidate at UIUC🌽; Formerly @GoogleDeepmind @NVIDIAAI @AmazonScience. #NLProc Large Language ModelsSubodh @subodhsharma
224 Followers 236 Following Asst. Prof. @ IIT Delhi • Formal Verification, Concurrency, Security • Retweets ≠ endorsementsYoungseok Yoon @YoungseokY87156
5 Followers 7 FollowingTian Qiu @QiuTian_QT
17 Followers 82 FollowingSharon Levy @sharonlevy21
1K Followers 573 Following Incoming Assistant Professor @RutgersCS Fall 2024 | Current Postdoc @jhuclsp PhD from UCSB @ucsbNLP Research in Responsible NLPZY Xue @Zhiyu_Xue228
27 Followers 455 FollowingDeepak Nathani @deepaknathani11
410 Followers 804 Following PhD Student @UCSBNLP | Mentor @CAREMentorship | Previously: Intern @AWS AI, Pre-Doctoral Researcher @GoogleAI India, Undergrad @IITHyderabadKexun Zhang @kexun_zhang
681 Followers 554 Following PhD student at @LTIatCMU. Previously at @ucsbNLP, @ZJU_china. language lover.Zekun Li @ZekunLi0323
760 Followers 587 Following CS Ph.D. student @UCSB, ex-intern at #MSFTResearch, Meta #RealityLabs, interested in #NLProc, #LLM, and #AI4HealthShayne Longpre @ShayneRedford
4K Followers 997 Following PhD @MIT. Prev: @Google Brain, @apple ML, @stanfordnlp. 🇨🇦 Interests: AI/ML/NLP, Data-centric AI, transparency & societal impactMadhav Ranka @mady619
177 Followers 160 Following Building @FincentHQ | Ex Engineering - @PayUindia | IIT Delhi '19 | Travel, Tech and LifeAman Rai @amankrai28
169 Followers 211 Following Learning delayed gratification (aka running a startup). Co-Founder at AroLeapAlon Albalak @AlbalakAlon
888 Followers 465 Following CS PhD candidate at @ucsbNLP. Research: Data-centric AI, Efficiency in ML, NLP.Safiya Umoja Noble @safiyanoble
66K Followers 17K Following 2021 #MacFellow 🥰 | Author, Algorithms of Oppression, @[email protected] & @safiyanoble.bsky.socialWenhu Chen @WenhuChen
11K Followers 520 Following AI researcher @UWaterloo @GoogleAI @VectorInst. Interested in natural language processing, diffusion models. I direct TIGER-Lab at UWaterloo.Danish Pruthi @danish037
7K Followers 628 Following Faculty at Indian Institute of Science, Bangalore. PhD from @LTIatCMU.Wenda Xu @WendaXu2
679 Followers 319 Following PHD candidate at UCSB’s NLP lab coadvised by William Wang and Lei LiAnthropic @AnthropicAI
262K Followers 26 Following We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant Claude at https://t.co/aRbQ97uk4d.UCSD CSE @ucsd_cse
3K Followers 392 Following Official Twitter page of the UC San Diego Department of Computer Science and EngineeringAK @_akhaliq
310K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo follow on Hugging Face: https://t.co/q2Qoey80GxBodhisattwa Majumder @mbodhisattwa
1K Followers 795 Following Research @allen_ai. Scientific Discovery, Language & Interactive Agents. PhD @ucsd_cse, @AdobeResearch Fellow. Prev @googleai @metaai @msftresearchJulien Chaumond @julien_c
47K Followers 1K Following Co-founder and CTO at @huggingface 🤗. ML/AI for everyone, building products to propel communities fwd. @Stanford + @PolytechniqueAntonis Antoniades @anton_iades
392 Followers 767 Following CS PhD student @ucsbNLP teaching machines to think like humans and humans to think like machines. Prev. @UCSBPhysics Guitar/Bouzouki. Cyprus/CA 🧠🤖🎸🌊I'm in the Baltimore/DC area this week to share some cool stuff we've been cooking! I'll give this talk at UMBC on Monday, Georgetown on Tuesday, UMD on Wednesday, and as a poster at MASC at JHU on Friday. If you're around and want to chat please ask! Happy to come say hi! 😃
Great piece by @matteo_wong in the @TheAtlantic quoting Dr. @IfeAdebara, who did her PhD with me at UBC Linguistics (@UBCLinguistics). Also quotes a number of fiends and colleagues, including @davlanade, @bonadossou @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1.…
AI understands just a tiny fraction of the thousands of languages used worldwide. That means the promised chatbot revolution could shut out billions of people, and even push some languages to extinction, @matteo_wong writes: theatlantic.com/technology/arc…
fixed the fixed fix for llama3
Thank you for the interview @matteo_wong @TheAtlantic , great to hear diverse opinions about the topic @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1
AI understands just a tiny fraction of the thousands of languages used worldwide. That means the promised chatbot revolution could shut out billions of people, and even push some languages to extinction, @matteo_wong writes: theatlantic.com/technology/arc…
AI understands just a tiny fraction of the thousands of languages used worldwide. That means the promised chatbot revolution could shut out billions of people, and even push some languages to extinction, @matteo_wong writes: theatlantic.com/technology/arc…
Brilliant piece @matteo_wong. FYA @nxthompson this is a good one! I do find the argument compelling & brilliant contributions from @davlanade @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1.
Thank you for the interview @matteo_wong @TheAtlantic , great to hear diverse opinions about the topic @bonadossou @IfeAdebara @anas_ant @alexandrabirch1 @sarahookr @HolyLovenia @mehakdh1
[ #Recherche ] 👩🔬 A l’occasion du "Mois #FemmesetScience", rencontre avec Nina Miolane, professeure d'intelligence artificielle à the University of California, Santa Barbara, où elle dirige le Geometric Intelligence Lab. ➡ bit.ly/3Ikwktb #IPPariswomeninscience
Who among us hasn't woken up in a cold sweat, terrified that Nicholas Carlini has demolished our security model overnight and uploaded the proof to arxiv?
Google announces Stealing Part of a Production Language Model We introduce the first model-stealing attack that extracts precise, nontrivial information from black-box production language models like OpenAI's ChatGPT or Google's PaLM-2. Specifically, our attack recovers the…
🚀 Exciting News! 🚀 We had our first Real AI group meeting last week, and it was a fantastic start to our journey together! It was amazing to see everyone's enthusiasm and passion for AI. At Real AI, we seek to pioneer the next frontiers of artificial intelligence for science.
🚀Fire linguists in the LLM era? No! Excited to share LingoLLM, a novel method for processing endangered languages with linguistic resources. LingoLLM improves the translation of many endangered languages from 0 to 10.5 BLEU! It helps other tasks as well! arxiv.org/abs/2402.18025
⚠️ Jailbreaking attacks for LLMs are crazy. How should we efficiently defend them? Check out 🛡️𝕊𝕒𝕗𝕖𝔻𝕖𝕔𝕠𝕕𝕚𝕟𝕘, a simple inference-time defense method. We found that fine-tuning with more safety data may not work well in defending wild jailbreaking attacks, while make…
another tutorial coming up built during the flight, this time DPO'ing Gemma 🥹
internet barely works in the plane, gg although I was able to train, perplexity was very bad still, PEFT + TRL is very fast 🥹💗
🎉 Thrilled to share my first first-author paper on adversarial transfer learning got accepted at @CVPR ! Feel free to check it out: arxiv.org/abs/2312.05716 Brief Summary ⏬
We identify a fundamental implicit bias of training neural networks with gradient descent. The weight matrices become aligned with a quantity closely related to the empirical NTK. This alignment is a mechanism for learning features from data.
Gradient descent induces alignment between weights and the empirical NTK for deep non-linear networks ift.tt/PZXqygl
TFW you ask ChatGPT a question and think ‘well that cant be the only right answer…’ In new work we look into how LLM alignment impacts pluralism and ability to reflect diverse opinions (it decreases it), and make an argument for pluralistic evaluations! arxiv.org/abs/2402.05070
🤔How can we align AI systems/LLMs 🤖 to better represent diverse human values and perspectives?💡🌍 We outline a roadmap to pluralistic alignment with concrete definitions for how AI systems and benchmarks can be pluralistic! arxiv.org/abs/2402.05070 First, models can be…
What are "representational harms"? This work: 1) examines current definitions 2) proposes an expansion to include affective and cognitive states 3) provides high-level requirements for measurement 4) outlines the harms wrt LLMs and 5) mitigation strategies arxiv.org/abs/2402.01705
This is the 7th year after my Ph.D. , and now looking back at my grad school, there are 5 things I wish I had known when I started graduate school. 🧵 1/6
I love to see ethics being brought into so many more of these conversations 🥰 @ucsd_cse #CSEOpenHouse2024
fun newsletter today on a recent pre-print suggesting AI models could be swamped by terrible machine-translated content in less common languages, with input from @emollick: politico.com/newsletters/di…