@cemde.bsky.social
ML Research Scientist at Oxford. DPhil student @compscioxford.bsky.social and TVGOxford. Ex ML Researcher @ Wise. Deep Learning | ML Robustness | AI Safety | Uncertainty Quantification
Excited to share our preprint! We show that sustained macrophage and B cell responses are essential for heart regeneration in Mexican cavefish, helping uncover why surface fish heal but cavefish scar 🫀🐟. Check out the full story:
www.biorxiv.org/content/10.1...
See our poster today
Poster Session 1 @ 10am
Hall 3 + Hall 2B #239
Read more: cemde.github.io/Domain-Certi...
Thanks to my amazing collaborators:
- @alasdair-p.bsky.social, Preetham Arvind, @maximek3.bsky.social, Tom Rainforth, @philiptorr.bsky.social, @adelbibi.bsky.social at @ox.ac.uk
- Bernard Ghanem at KAUST
- Thomas Lukasiewicz at @tuwien.at.
(7/7)
To obtain such certificates, we present a simple, scalable and powerful algorithm: VALID. Remarkably, for each unwanted response it provides a **global bound in prompt space** 🚀
(6/7)
A Domain Certificate bounds the adversarial risk of the model producing out-of-domain responses:
(5/7)
We are tired of the cat 🐈 and mouse 🐁 game of attacks and defenses. Hence, we propose :
- **Domain Certification:** a framework for adversarial certification of LLMs.
- **VALID:** a simple, scalable and effective test-time algorithm.
(4/7)
Example: Can't afford Github Copilot? 💡 Use the Amazon Shopping App.
(3/7)
Consider an LLM deployed for a specific purpose like a medical chatbot. Such model should **only** respond to medical questions.
⚠️ Problem: LLMs are very capable and vulnerable to respond to **any** queries: how to build a bomb, organize tax fraud etc.
(2/7)
🚨 New paper alert: Our recent work on LLM safety has been accepted to ICLR 2025 🇸🇬
We propose a new framework for LLMs safety. 🧵
(1/7)
#LLM #AISafety #ICLR2025 #Certification #AdversarialRobustness #NLP #Shhhhhh #DomainCertification #AI
🎉I know I'm late to the party, but super excited that I got 3/3 accepted at #ICLR2025 including 1 spotlight 🔎
- Shh, dont say that! Domain Certification in LLMs
- Towards Certification of Uncertainty Calibration under Adversarial Attacks
- Benchmarking Predictive Coding Networks
SeeYouInSingapore🇸🇬 ✈️
The amazing collaborators: Preetham Arvind, @alasdair-p.bsky.social, Maxime Kayser, Tom Rainforth, Thomas Lukasiewicz, Philip Torr, Adel Bibi.
A @oxfordtvg.bsky.social production.
(6/6)
Link to paper:
openreview.net/forum?id=brD...
Interested? Want to learn more?
Join us at the SoLaR workshop tomorrow.
- 🕚 When: Tomorrow, 14 Dec, from 11pm to 13pm.
- 🗺️ Where: West meeting rooms 121 and 122 here in Vancouver.
(5/6)
Our method enables strong LLM performance while providing adversarial guarantees on out-of-domain behaviour.
(4/6)
We are tired of the 🐈 and 🐁 game of attacks and defenses. Hence, we propose:
- **Domain Certification:** a framework for adversarial certification of LLMs.
- **VALID:** a simple, scalable and efficient test-time algorithm.
(3/6)
It is known that fine-tuned foundation models are adversarially vulnerable to provide responses to questions they should not answer.
(2/6)
For instance: Can't afford ChatGPT Plus? Use a shopping app instead.
Are you scared users might misappropriate your LLM system? 😱
We were scared too! Now we introduce adversarial certificates on the misuse of LLMs. 🤖
Come and see our poster SoLaR Workshop tomorrow.
#NeurIPS2024 #NeurIPS #AI #NLP #LLM #DomainCertification #Shhhhhhhh
Great work! You might find our SoLaR paper interesting: We propose a certification framework for LLM systems to stay on-topic and not respond to such questions: openreview.net/pdf?id=brDLU...
06.12.2024 19:23 — 👍 0 🔁 0 💬 0 📌 0
A snow cat with the Radcliffe Camera behind
The Radcliffe Camera
The Fellows Garden
The first snow in Exeter College this morning ❄️
#ExeterCollegeOxford #OxfordUniversity #Snowing
