Search papers, labs, and topics across Lattice.
University of Alberta
2
0
4
8
Rollout activation quantization, not training-side errors, is the hidden culprit behind FP4 RL accuracy degradation, and a clever residual correction can bridge the gap to full precision.
By co-training flow and retrieval networks, WINFlowNets eliminates the need for pre-training, unlocking CFlowNets for dynamic robotic environments where data is scarce.
