Re:Real-Time Introspective Compression for Transformers
by pumpikano on Wednesday April 02, 2025 @03:25PM
(#43560534)
Cool! On a quick glance, it doesn't seem like the group/layer index is provided to the compression model. That might help a bit with fidelity at pretty low additional cost.
[ Parent
| on HN ]
Re:Real-Time Introspective Compression for Transformers
by eigenvalue on Wednesday April 02, 2025 @03:43PM
(#43560705)
Interesting, it certainly wouldn’t take up much additional space, but I wonder if it would have any real impact, since it seems somewhat orthogonal to finding a faithful low-dimensional encoding of the activations.