Recently, large language models (LLMs) have shown surprising performance in
task-specific workloads as well as general tasks with the given prompts.
However, to achieve unprecedented performance, recent LLMs use billions to
trillions of parameters, which hinder the wide adaptation of those models due
to their extremely large compute and memory requirements. To resolve the issue,
various model compression methods are being actively investigated. In this
work, we propose SDQ (Sparse Decomposed Quantization) to exploit both
structured sparsity and quantization to achieve both high compute and memory
efficiency. From our evaluations, we observe that SDQ can achieve 4x effective
compute throughput with <1% quality drop.

借助 SDQ 的结构稀疏性和量化方法，可以实现高计算和内存效率，并且在性能上只有不到 1% 的质量损失。

SDQ：稀疏分解量化用于 LLM 推断

SDQ: Sparse Decomposed Quantization for LLM Inference

Doodling is a useful and common intelligent skill that people can learn and
master. In this work, we propose a two-stage learning framework to teach a
machine to doodle in a simulated painting environment via Stroke Demonstration
and deep Q-learning (SDQ). The developed system, Doodle-SDQ, generates a
sequence of pen actions to reproduce a reference drawing and mimics the
behavior of human painters. In the first stage, it learns to draw simple
strokes by imitating in supervised fashion from a set of strokeaction pairs
collected from artist paintings. In the second stage, it is challenged to draw
real and more complex doodles without ground truth actions; thus, it is trained
with Qlearning. Our experiments confirm that (1) doodling can be learned
without direct stepby- step action supervision and (2) pretraining with stroke
demonstration via supervised learning is important to improve performance. We
further show that Doodle-SDQ is effective at producing plausible drawings in
different media types, including sketch and watercolor.

通过 Stroke Demonstration 和 deep Q-learning (SDQ) 提出了 Doodle-SDQ 两阶段学习框架，该系统可以产生一系列的笔触行为来复制参考图，并模仿人类画家的行为。实验结果证实了，(1) 在没有直接的逐步行动监督的情况下可以学会 doodling 的技能；(2) Stroke Demonstration 通过监督学习的预训练对于提高性能是很重要的。同时 Doodle-SDQ 还可以有效地在不同类型的媒介中产生逼真的图像，包括草图和水彩画。