Sessions
| Date | Presenter | Topic/Paper |
|---|---|---|
| 18th Aug’25 | Hardik Jindal | Compositional Exemplars for In-Context Learning |
| 18th Aug’25 | Himanshu Sharma | Sparse AutoEncoders find highly interpretable features in language models |
| 18th Aug’25 | Parth Patil | Understanding the Disharmony between Dropout and Batch Normalization by Variance Shift |
| 18th Aug’25 | Pratik Kumar | Optimizing ML training with Metagradient |
| 25th May’25 | Aarush Singh Kushwaha | Reinforcement Learning for Reasoning in LLMs |
| 6th Apr’25 | Himanshu Sharma | Mechanistic Interpretability: Finding Neurons in a Haystack: Case Studies with Sparse Probing |
| 23rd Mar’25 | Malgatte Harsh Siddharth | Attention is All You Need |
Attending the sessions and presenting papers/ideas
- If you would like to present a paper to the group, please fill the given form
- Information regarding presentations and group meetings shall be disseminated via machine-learning-interest-group channel of the Programming Club Discord Server (join via the invite link ). The server is only accessable to members who have the machine-learning-interest-group role. This can be obtained via heading towards #role-claim channel, and reacting to the MEE6 message with a 🦝 emoji.