Vision-based end-to-end driving models trained by imitation learning can lead
to affordable solutions for autonomous driving. However, training these
well-performing models usually requires a huge amount of data, while still
lacking explicit and intuitive activation maps to reveal the inner workings of
these models while driving. In this paper, we study how to guide the attention
of these models to improve their driving quality and obtain more intuitive
activation maps by adding a loss term during training using salient semantic
maps. In contrast to previous work, our method does not require these salient
semantic maps to be available during testing time, as well as removing the need
to modify the model's architecture to which it is applied. We perform tests
using perfect and noisy salient semantic maps with encouraging results in both,
the latter of which is inspired by possible errors encountered with real data.
Using CIL++ as a representative state-of-the-art model and the CARLA simulator
with its standard benchmarks, we conduct experiments that show the
effectiveness of our method in training better autonomous driving models,
especially when data and computational resources are scarce.

通过使用显著语义地图进行训练期间添加损失项，引导注意力可以改善自动驾驶模型的驾驶质量，并获得更直观的激活图，特别是在数据和计算资源有限的情况下。

引导端到端驾驶模型的注意力

Guiding Attention in End-to-End Driving Models

Denoising Diffusion models have exhibited remarkable capabilities in image
generation. However, generating high-quality samples requires a large number of
iterations. Knowledge distillation for diffusion models is an effective method
to address this limitation with a shortened sampling process but causes
degraded generative quality. Based on our analysis with bias-variance
decomposition and experimental observations, we attribute the degradation to
the spatial fitting error occurring in the training of both the teacher and
student model. Accordingly, we propose $\textbf{S}$patial
$\textbf{F}$itting-$\textbf{E}$rror $\textbf{R}$eduction
$\textbf{D}$istillation model ($\textbf{SFERD}$). SFERD utilizes attention
guidance from the teacher model and a designed semantic gradient predictor to
reduce the student's fitting error. Empirically, our proposed model facilitates
high-quality sample generation in a few function evaluations. We achieve an FID
of 5.31 on CIFAR-10 and 9.39 on ImageNet 64$\times$64 with only one step,
outperforming existing diffusion methods. Our study provides a new perspective
on diffusion distillation by highlighting the intrinsic denoising ability of
models.

通过分析偏差 - 方差分解和实验观察，我们提出了一种基于空间拟合误差减少的蒸馏模型（SFERD），该模型利用教师模型的注意力引导和设计的语义梯度预测器来降低学生模型的拟合误差，从而在少量函数评价中实现高质量的样本生成，并在 CIFAR-10 上取得了 5.31 的 FID 和 ImageNet 64×64 上取得了 9.39 的 FID，超过了现有的扩散方法。我们的研究突出了模型的本质去噪能力，为扩散蒸馏提供了新的视角。

减少噪声扩散模型蒸馏中的空间拟合误差

Reducing Spatial Fitting Error in Distillation of Denoising Diffusion  Models

Augmented reality (AR) games are a rich environment for researching and
testing computational systems that provide subtle user guidance and training.
In particular computer systems that aim to augment a user's situation awareness
benefit from the range of sensors and computing power available in AR headsets.
In this work-in-progress paper, we present a new environment for research into
situation awareness and attention guidance (SAAG): an augmented reality version
of the board game Carcassonne. We also present our initial work in producing a
SAAG pipeline, including the creation of game state encodings, the development
and training of a gameplay AI, and the design of situation modelling and gaze
tracking systems.

本论文介绍了使用 AR 技术的游戏来进行情境感知和关注引导 (SAAG) 的研究，并介绍了游戏状态编码、游戏 AI 的开发与训练以及情境模型和凝视追踪系统的设计。