Reposted by Jia ▫️Mingqian Zheng @mingqian-zheng.bsky.social · 20/10/2025How and when should LLM guardrails be deployed to balance safety and user experience? Our #EMNLP2025 paper reveals that crafting thoughtful refusals rather than detecting intent is the key to human-centered AI safety. 📄 arxiv.org/abs/2506.00195 🧵[1/9] 193
Reposted by Jia ▫️Mingqian Zheng @mingqian-zheng.bsky.social · 20/10/2025[9/9] Big THANKS to my amazing collaborators @jiajiah.bsky.social @pigeonzow.bsky.social Motahhare Eslami, Jena Hwang @faebrahman.bsky.social, @carolynrose.bsky.social @maartensap.bsky.social from @ltiatcmu.bsky.social Pareto.ai @sfu.ca @ai2.bsky.social ♥️ 📂 github.com/EEElisa/LLM-Guardrailsgithub.comGitHub - EEElisa/LLM-GuardrailsContribute to EEElisa/LLM-Guardrails development by creating an account on GitHub. 022
Jia ▫️ @jiajiah.bsky.social · 02/06/2025Some life updates here: got a car, managed to escape grad school, working on expert data curation and future of work these days, had a new cat, moved back to SF, finally feels alive now. 050
Reposted by Jia ▫️Patrick @ninjapleasedj.bsky.social · 19/05/2023Ok, guys I cleared out the fridge and I… OH NO 1923442
Jia ▫️ @jiajiah.bsky.social · 05/05/2023From what I’ve see, most proposed LLM based tool is really just as a proof of concept. If all the “moderation tool” do is to prompt the GPT to produce a formatted JSON file, there’s no way to tune and no barriers of entry. 022
Jia ▫️ @jiajiah.bsky.social · 04/05/2023My daily reminder of Crazy people posts, but most people are neither crazy nor post. 010
Reposted by Jia ▫️Jay 🦋 @jay.bsky.team · 03/05/2023it’s important for y’all to realize that this app is wild right now because all of you are posters. you’re probably not meant to meet each other in the wild without a cushioning biome of lurkers around 52642129
Jia ▫️ @jiajiah.bsky.social · 04/05/2023[1/] My immediate world has been crazy lately, centralized and decentralized social media, generative AI, political ideology getting extreme. Making sense of those madness makes me feel vulnerable and disoriented. 120
Reposted by Jia ▫️Rose 🌹 @rose.bsky.team · 27/04/2023Bluesky SF Meetup #2 happening next Wed (5/3) at the SF Commons. @emily.bsky.team and I will be there to hangout, answer q's, handout invites, etc. Pls RSVP 💙🦋: lu.ma/54jtbp4n 2711223
Jia ▫️ @jiajiah.bsky.social · 27/04/2023Some how @jaz.bsky.social’s post interacted based social graph really incentivize me to post more. 030
Reposted by Jia ▫️Jay 🦋 @jay.bsky.team · 28/01/2023STEAL HER LOOK: - lace blouse by chanel - straw boater hat by dior - engraved locket with your true love’s photo - an obituary that reads “she remained a spinster, sharing a room with her best friend until death did they part” 511110
Jia ▫️ @jiajiah.bsky.social · 26/04/2023I’ve already been curious about the question “can vibe scale”. It often goes small community has good vibe -> people rush in to experience the vibe through consumption rather than creation -> social signal needed for identify who creates the vibe. 050
Jia ▫️ @jiajiah.bsky.social · 25/04/2023Does a healthy community self regulate the same way a healthy person does? Playing infinite games, pursue balance and wholesomeness, have trust ? 120
Jia ▫️ @jiajiah.bsky.social · 25/04/2023What does it take for a community to remain healthy and self regulate? 010
Jia ▫️ @jiajiah.bsky.social · 25/04/2023co pilot really super charged my coding skills! I didn’t realize how much dyslexia has make it extremely difficult for me to code until I start to use co-pilot. 010
Reposted by Jia ▫️Peter W🤓 @wang.social · 25/04/2023I forget where I heard the term, but someone mentioned to me about adopting the moniker “web0” to refer to this reset. I can’t wait to see a At protocol implementation in Pyscript and get truly serverless social app fabric in the browser/client. 1406
Jia ▫️ @jiajiah.bsky.social · 25/04/2023totally agree! Spend sometimes reading the documentation today and it really felt that bsky is aligned with the early internet ethos from the design of protocols and up. 000
Jia ▫️ @jiajiah.bsky.social · 24/04/2023The api tooling seems like a helpful repo for me to figure out how to do data collection 010
Reposted by Jia ▫️Louis @louis.bsky.social · 22/04/2023ATProtocol Ecosystemにぽーまんさんが作ったクライアント「Skylight」を僕がカスタマイズした「Skylight Bluesky Style」もあって良かった。プルリクエストをした甲斐があった。 bluesky-social/atproto-ecosystem: list of projects and implementations in the AT protocol ecosystem github.com/bluesky-social/atproto-e… 051
Jia ▫️ @jiajiah.bsky.social · 22/04/2023Community curation is really more a intricate art than science. All about vibe building. 020
Jia ▫️ @jiajiah.bsky.social · 22/04/2023Just realize bsky still have the nice wholesome village vibe despite there’s many people from different countries hanging out here now. 020
Reposted by Jia ▫️평온한 야생사슴 ✝️🌈👶 @osssneru.bsky.social · 21/04/2023이 기능들 정말 좋다 직관적인 이미지 필터링 기준 두가지, 자극적인 요소 필터링, 그리고 혐오그룹 필터링과 스팸 및 불량이용자 등등 필터링까지 다 있음 그리고 앞으로 이런 설정들이 세세해지고 그걸 개발자들이 지지해줄거란 믿음이 있는게 좋다 036
