DIET: Deletion-response Expert Trimming for Video Diffusion Transformers
Quick summary
arXiv:2609.37829v1 Announce Type: new Abstract: Video diffusion transformers (DiTs) increasingly adopt mixture-of-experts (MoE) architectures to reduce active computation, but their full expert storage remains costly. Existing one-shot pruning criteria mainly rely on static activation or routing statistics and cannot capture layer-level re-routing after expert deletion. We introduce DIET, a training-free expert pruning framework based on deletion responses. A single all-expert calibration pass records expert outputs and router states for matched conditional and unconditional tokens. Candidate
Key takeaways
- arXiv:2609.37829v1 Announce Type: new Abstract: Video diffusion transformers (DiTs) increasingly adopt mixture-of-experts (MoE) architectures to reduce active computation, but their full expert storage remains costly.
- Existing one-shot pruning criteria mainly rely on static activation or routing statistics and cannot capture layer-level re-routing after expert deletion.
- We introduce DIET, a training-free expert pruning framework based on deletion responses.
Why it matters
“DIET: Deletion-response Expert Trimming for Video Diffusion Transformers” illustrates how changes in the AI ecosystem can affect products, workflows and user expectations together. Its lasting significance depends on measurable adoption, cost and safety outcomes.
Member comments