DLoop: Looped Speculative Decoding
DLoop loops speculative decoding so draft models propose more tokens before one joint verification, cutting target-model passes.
52
DLoop loops speculative decoding so draft models propose more tokens before one joint verification, cutting target-model passes.
NP-OPD improves on-policy distillation by using weaker-model rollouts as an explicit negative signal.
Today 381 readers · 421 views · 38 reads · all time 109K views