We have 6 of 27 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
This work introduces Schur Replay, a scale-selection algorithm that reproduces the GPTQ updates caused by each block scale and scores the resulting block error after accounting for compensation from unquantized columns.
This work formalizes per-batch dispatch as a fixed-charge makespan problem---NP-hard on two fully replicated GPUs, polynomial in degenerate limits---and presents TEMPO, a makespan-aware dispatcher solving it in milliseconds off the critical path; its SGLang integration runs out-of-process and fuses dispatch with count...
JAGG approximates intermediate-step Jacobians via $t$-weighted interpolation of the endpoint Jacobians, then aggregates per-step upstream signals into two composite gradients applied through a single joint backward pass, and it is proved this interpolation is exact when the velocity is linear in $(z,t)$, and a cosine-s...
This work proposes Latent Reward Registers, a mechanism that estimates terminal preference directly from intermediate noisy latents, and achieves significant reward improvement with a favorable reward-quality balance against training-free baselines.
A lightweight Burst-Gap model parameterized by backpressure-free issue time, effective drain rate, and outstanding capacity predicts issue overhead, recovery between bursts, and the onset of backpressure, and two communication-computation fused kernels are redesigned.
HeadCast is proposed, a training-free, plug-and-play acceleration framework built on the observation that a pre-trained AR model's attention heads exhibit stable, heterogeneous behaviors that accelerates inference by up to 1.62x at 720P and 1.95x at 1080P, while keeping VBench quality on par with full attention and lar...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
