AI Research papers walkthrough with Akhil Sharma (online)

In the true spirit of 'R' in DARLing Meetup, we start the series of reading research papers and discussing them. A big thank you to Akhil Sharma to kickstart this. Special thanks to Kishore for converting thought into action.

Why read research papers?

• AI field is evolving fast and there are many universities, companies and individuals pushing the limits of what is possible. Most of the courses and YouTube videos cover AI at a superficial level.
• AI research papers give you the ability to see-through the algorithms, mathematics and understand the power. Initially, they appear intimidating. This series is an attempt to overcome that.

What to expect:

• This session is for people serious about AI/ML and those who are curious.
• Don't expect to understand everything, but come with an open mind to learn new things in research.
• This could trigger new thoughts and create interest for you to pursue your own research.

Research paper 1: Why Do Some Language Models Fake Alignment While Others Don’t?
Link: https://arxiv.org/pdf/2506.18032

Research paper 2: Inoculation Prompting: Instructing Llms To Misbehave At Train-time Improves Test-time Alignment
Link: https://arxiv.org/pdf/2510.05024

Session Facilitator: Kishore Vutukuri