TRAIL - Trustworthy and Responsible AI Lab

TRAIL is a Sorbonne University joint research lab to study and address challenges linked to a responsible and trustworthy use of AI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

🥳 Happy to announce that our work "Understanding prediction discrepancies in classification" has been accepted to Machine Learning and is now available: https://t.co/4EZB0vSepf

Also, @XavierRenard will be presenting this work next week at @ECMLPKDD #ExplainableAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Why do explanations fail?

In #XAI, many issues may hurt user trust. As current studies often remain ad-hoc, @ClaraZiemann, @thibaultlaugel @detyniecki et al.propose in this work a typology of XAI failures capturing both technical and user-based failures:
https://t.co/B6XqHfFoOH

[Thibault Laugel @thibaultlaugel]

Attending #ICLR2024 ? Join @GrariVincent and I this afternoon at 4:30pm for our poster session!

In our work, we propose a new method, ROAD, to build models that are not only fair globally, but locally as well!

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

🚨🚨 2 NEW OPEN POSITIONS (#PhD Student) 🚨🚨
We are looking for our next batch of PhD students to advance research on #LLM!
1) Trustworthy LLM with Natural Language Instructions ⚖
2) Evaluation of Reasoning in LLM 🧠

More info: https://t.co/J8lM41xkjs

Come work with us!

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢 PhD Defense Announcement 📢

Next Tuesday afternoon, @AdulamJeyasothy will be defending her PhD "Generating personalized post-hoc explanations" at @LIP6_lab!

You would like to attend? Send us a DM to have the information.
#XAI #ResponsibleAI

[Thibault Laugel @thibaultlaugel]

🥳🥳🥳Our work on algorithmic fairness generalization has been accepted to #ICLR2024! @trail_lab

Written with @GrariVincent @tatsu_hashimoto @SLamprier @detyniecki during my visiting stay @CodeXStanford ❤️

Paper: https://t.co/WA1W3ouSwX
Some details below 👇

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢The next TRAIL seminar is on Thursday! Do not forget to register to attend 👇

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Exciting TRAIL Seminar announcement! 🚨🚨🚨

On November 16, we'll be welcoming 3 PhD students to hear their different perspectives on #ResponsibleAI

Information and inscription (on-site and virtual) here: https://t.co/s5QKGyGFQ5

Details on each presentation below!👇 (1/4)

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

@telecomparis @IP_Paris_ @LIP6_lab @Sorbonne_Univ_ @ekimetrics (4/4)
Thomas Souverain is a PhD student in #Philosophy @ENS_ULM, working with @DreamQuark and @poleemploi_IDF

In his presentation, he will discuss the alignment between the perceived values of justice and transparency and their implementations in the field of #ResponsibleAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Exciting TRAIL Seminar announcement! 🚨🚨🚨

On November 16, we'll be welcoming 3 PhD students to hear their different perspectives on #ResponsibleAI

Information and inscription (on-site and virtual) here: https://t.co/s5QKGyGFQ5

Details on each presentation below!👇 (1/4)

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

@telecomparis @IP_Paris_ (3/4) Milan Bhan is a PhD Student
@LIP6_lab @Sorbonne_Univ_ and @ekimetrics , working on #XAI

In his presentation "Generating textual counterfactual explanations", he will discuss the current state of XAI in #NLP, and introduce new methods to make text models transparent

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Yesterday, at the TRAIL seminar, Pr. Bilel Benbouzid from @UGustaveEiffel and #LISIS lab presented his work on "Politics of AI". An insightful talk, about the societal impacts of #AI through a political prism, which sparked lots of discussions! 💡 #AIEthics #STS

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Tune-in tomorrow for the ACRAI @UAntwerpen launch event! featuring @XavierRenard with a keynote: "When Responsible AI Research Needs to Meet Reality" #XAI #ResponsibleAI

[Applied Data Mining @ApplDataMining]

🚀Launching 𝐀𝐂𝐑𝐀𝐈: Antwerp's hub for Responsible AI! 📅28th Sept. Highlights: 🕔Intro & TRAIL collab 🕠ChatGPT session w/ Prof. Daelemans 🕕Keynote by Mr. Renard @ AXA 🕡Meet the ACRAI team 🕖Reception 📍Antwerp or online. RSVP👉 https://t.co/G71YA5XmUu🚀 #ResponsibleAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

Save the date for the next TRAIL seminar! 📅
On Oct. 12th, Bilel Benbouzid will present his work (in French) on AI policy. Information and registration: https://t.co/9fpEQNi02b
Attendance is possible both on-site and online, pending registration.
#XAI #ResponsibleAI #AIpolicy

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

🎉🍾 And that's a wrap! Congratulations to @ClaraZiemann who successfully defended her PhD on Friday @LIP6_lab @Sorbonne_Univ_ in front of a convinced jury @katrien_v @Quecalcoatle @detyniecki Nicolas Labroche and Theodoros Evgeniou! #XAI #ResponsibleAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢 @ClaraZiemann 's PhD Defense is about to start!

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

During her thesis (under the supervision of M.-J. Lesot, C. Tijus and @detyniecki), @ClaraZiemann focused on XAI, and how to improve the intelligibility of explanations to help non-expert users understand AI decisions. Interested in learning more? Reach out!

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢🎉 On Friday, @ClaraZiemann will be defending her PhD thesis "Designing and evaluating explanation user interfaces for complex machine learning systems" at @Sorbonne_Univ_ @LIP6_lab !

More information and how to attend: https://t.co/DXFgXtKQkJ

#XAI #ResponsibleAI #phdthesis

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢Tune in for the launch event of our partner center ACRAI from @UAntwerpen! Featuring a keynote from our very own @XavierRenard!
Researchers from TRAIL and ACRAI have already been working together for several years. Looking forward to the future of this collaboration! 🤝

[Applied Data Mining @ApplDataMining]

🚀Launching 𝐀𝐂𝐑𝐀𝐈: Antwerp's hub for Responsible AI! 📅28th Sept. Highlights: 🕔Intro & TRAIL collab 🕠ChatGPT session w/ Prof. Daelemans 🕕Keynote by Mr. Renard @ AXA 🕡Meet the ACRAI team 🕖Reception 📍Antwerp or online. RSVP👉 https://t.co/G71YA5XmUu🚀 #ResponsibleAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢 This Thursday - 2pm -
@Sorbonne_Univ_ Jussieu campus at @LIP6_lab or online (follow the link)

6th TRAIL Seminar on -Large Language Models like GPT-4/ChatGPT in the practice of law-

by @HarrySurden, Prof. of Law at the Univ. of Colorado

#LLM #LLMs #ResponsibleAI

[TRAIL - Trustworthy and Responsible AI Lab @trail_lab]

📢 TRAIL Seminar - June 29th at @ScienceSorbonne and @LIP6_lab

➡️ Talk on Large Language Models & Law

by @HarrySurden, Prof. of Law at the Univ. of Colorado @ColoLaw & Associate Director @CodeXStanford

Details + Registration 🔎https://t.co/cVFJ16GWY2

#LLM #LLMs #ResponsibleAI