// PERSON
Varun Giridhar
MENTIONS 1LAST SEEN AUGUST 21, 2026
// RECENT MENTIONS
// SIGNALS
1 SIGNAL
01
mention·arXiv Physical AI·AUGUST 21, 2026
“The core breakthrough of this paper is exploiting the asymmetry between behavior cloning (BC) and Q-learning: BC can only be trained on successful demonstrations, while an off-policy Q-function can be trained on *any* rollout, including failures.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.