π₯ New Paper!
Models are trained not to change their answers just because a user disagrees.
But when the same wrong claim comes from a βverified source,β they give in.
We call this Authority Bias.
Accepted at NeurIPS 2026.
π§΅
π’ Releasing our latest paper (Accepted AAAI 26)
Students have immense research potential, but enough mentors for them. What if we could design an AI system to mentor them?
We introduce METIS (Mentoring Engine for Thoughtful Inquiry & Solutions), a stage-aware research mentor.
We attended the India Game Dev Conference (IGDC) recently. It was a good opportunity to meet industry professionals, learn about new trends, and connect with other developers.
Meet Karan (@karan_pargal), McCarthy Lab alumnus. Initially focused on ML engineering, a hackathon pivoted him to Web3, where heβs earned significant recognition.
took a talk on 'how to do effective research' @nexttechlab, really loved the crowd turnout.
Majority of the topics I discussed were part of the learnings I gained during my internship at @UofIllinois & @lossfunk
Link to the resource in comments.
Thank You, @nexttechlab and everyone I met along my journey at the lab.
4 years at UB 706 prepared me set things in motion at my first day, interning at @iiscbangalore
As he gradually built up his skills in IT, Bryan joined @Boeing in 2019 where he's responsible for software and hardware integration of flight simulation devices.