Kushal Chakrabarti
PulseAugur coverage of Kushal Chakrabarti — every cluster mentioning Kushal Chakrabarti across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New ERASE framework enables AI model forgetting without weight changes
Researchers have introduced ERASE, a novel framework designed for on-the-go forgetting of private data within AI models without altering their weights. This method utilizes adversarial signal editing and class-condition…
-
New distributed optimization method uses energy conservation laws
A new research paper introduces a novel approach to distributed optimization using energy conservation laws within dilated coordinates. The proposed method achieves an O(t^{-2}) rate for objective gap and consensus erro…
-
Bigger AI models compound mistakes faster, new research finds
A new research paper from Kushal Chakrabarti highlights a concerning trend in large language models: as they scale, their ability to generate correct answers decreases faster than previously understood. This phenomenon,…