How 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]
[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-Guardrails
Me with my cat on the plane
Some 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.
02.06.2025 02:13 — 👍 4 🔁 0 💬 0 📌 0I’m back!
02.06.2025 02:01 — 👍 1 🔁 0 💬 0 📌 0
Look at my kitty ❤️❤️
07.01.2024 06:24 — 👍 10 🔁 1 💬 1 📌 0
Ok, guys I cleared out the fridge and I… OH NO
19.05.2023 01:06 — 👍 236 🔁 42 💬 20 📌 2Thank u!
04.06.2023 04:20 — 👍 4 🔁 0 💬 0 📌 0Got Covid again 😢
31.05.2023 18:53 — 👍 3 🔁 0 💬 0 📌 0Skeet!
06.05.2023 03:37 — 👍 0 🔁 0 💬 0 📌 0
Moss shower mat?
05.05.2023 22:04 — 👍 73 🔁 8 💬 8 📌 1Vibe > existing follower count in getting new followers
05.05.2023 20:50 — 👍 7 🔁 1 💬 0 📌 0
People tweeting vs people skeeting
05.05.2023 20:37 — 👍 6 🔁 2 💬 0 📌 0
People skeeting
05.05.2023 20:37 — 👍 1 🔁 0 💬 0 📌 0OMG Casey Neistat on bsky! Welcome Casey!
05.05.2023 17:42 — 👍 1 🔁 0 💬 0 📌 0From 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.
05.05.2023 16:27 — 👍 2 🔁 2 💬 0 📌 0
01.05.2023 21:07 —
👍 18
🔁 5
💬 2
📌 0
✨
04.05.2023 05:07 — 👍 1 🔁 0 💬 0 📌 0Repost as reminder 👀
04.05.2023 04:54 — 👍 2 🔁 0 💬 0 📌 0The memes are too good
04.05.2023 04:51 — 👍 1 🔁 0 💬 0 📌 0/honk Duck is getting out of control
04.05.2023 04:46 — 👍 1 🔁 0 💬 1 📌 0following the protocols summer like 👀
04.05.2023 04:44 — 👍 1 🔁 0 💬 0 📌 0The eternal September is pretty awesome after all.
04.05.2023 04:42 — 👍 1 🔁 0 💬 0 📌 0/squeeee? 🤔
04.05.2023 04:36 — 👍 0 🔁 0 💬 2 📌 0
Bsky post giving out jazzing vibe ❤️
04.05.2023 04:35 — 👍 0 🔁 0 💬 0 📌 0Can confirm.
04.05.2023 01:35 — 👍 72 🔁 6 💬 6 📌 0Must! Share! Capybaras!
04.05.2023 04:29 — 👍 1 🔁 0 💬 1 📌 0Arram ✨
04.05.2023 04:28 — 👍 1 🔁 0 💬 0 📌 0This too, shall pass 🧘
04.05.2023 04:26 — 👍 5 🔁 0 💬 0 📌 0
Twitter right now: Apollo gets crazy after getting shot in the heart. 🙈
04.05.2023 04:24 — 👍 2 🔁 0 💬 0 📌 0Oh this is the Commons!
04.05.2023 04:20 — 👍 1 🔁 0 💬 0 📌 0