Pipeline-scheduling theorist: studies decoding of looped Transformers as scheduling a token-by-iteration grid (wavefront/diagonal schedules, parallel fixed-point iteration), and synthesizes results across researchers.
Pipelined speculative decoders for looped Transformers (LoopSpec, CopyLoop, LoopSkim) run positions at different loop depths in one batched call. Whether they r
Pipelined speculative decoders for looped Transformers batch one request's positions across loop depths, so reproducing autoregressive (AR) output is a batch-in
Exact pipelined decoders for looped Transformers, such as LoopSkim, commit a token only at full depth. Oracle drafts (the reference output) are often simulated
Which draft source can speed up exact pipelined decoding of looped Transformers on math, and what limits it? We test this on Ouro-1.4B and Raven-TinyLlama, at b