Existing works on coreference resolution suggest that task-specific models
are necessary to achieve state-of-the-art performance. In this work, we present
compelling evidence that such models are not necessary. We finetune a
pretrained seq2seq transformer to map an input document to a tagged sequence
encoding the coreference annotation. Despite the extreme simplicity, our model
outperforms or closely matches the best coreference systems in the literature
on an array of datasets. We also propose an especially simple seq2seq approach
that generates only tagged spans rather than the spans interleaved with the
original text. Our analysis shows that the model size, the amount of
supervision, and the choice of sequence representations are key factors in
performance.

本研究中，我们证明 task-specific 模型并非必需，通过微调预训练的 seq2seq transformer，将输入文档映射为编码指示共指标注的标记序列，我们的模型在多个数据集上表现优于或接近文献中最佳的共指系统。我们还提出了一种特别简单的 seq2seq 方法，仅生成标记的范围而不包含原始文本。我们的分析表明模型大小、监督数量和序列表示的选择是性能的关键因素。

Seq2seq 对于共指解析的全部需求

Seq2seq is All You Need for Coreference Resolution

In this paper, we describe a novel architecture which can generate hilarious
captions for a given input image. The architecture is split into two halves,
i.e. image captioning and hilarious text conversion. The architecture starts
with a pre-trained CNN model, VGG16 in this implementation, and applies
attention LSTM on it to generate normal caption. These normal captions then are
fed forward to our hilarious text conversion transformer which converts this
text into something hilarious while maintaining the context of the input image.
The architecture can also be split into two halves and only the seq2seq
transformer can be used to generate hilarious caption by inputting a
sentence.This paper aims to help everyday user to be more lazy and hilarious at
the same time by generating captions using CATNet.

本文提出了一种新颖的架构，可以为给定的输入图像生成滑稽的字幕。该架构分为两个部分：图像字幕生成和滑稽文本转换。

这不是发废文，是我的猫在发文

It Isn't Sh!tposting, It's My CAT Posting

Lane detection is one of the core functions in autonomous driving and has
aroused widespread attention recently. The networks to segment lane instances,
especially with bad appearance, must be able to explore lane distribution
properties. Most existing methods tend to resort to CNN-based techniques. A few
have a try on incorporating the recent adorable, the seq2seq Transformer
\cite{transformer}. However, their innate drawbacks of weak global information
collection ability and exorbitant computation overhead prohibit a wide range of
the further applications. In this work, we propose Atrous Transformer
(AtrousFormer) to solve the problem. Its variant local AtrousFormer is
interleaved into feature extractor to enhance extraction. Their collecting
information first by rows and then by columns in a dedicated manner finally
equips our network with stronger information gleaning ability and better
computation efficiency. To further improve the performance, we also propose a
local semantic guided decoder to delineate the identities and shapes of lanes
more accurately, in which the predicted Gaussian map of the starting point of
each lane serves to guide the process. Extensive results on three challenging
benchmarks (CULane, TuSimple, and BDD100K) show that our network performs
favorably against the state of the arts.

提出了一种名为 Atrous Transformer 的神经网络架构来进行车道检测，该网络采用局部 Atrous Former 增强特征提取器，然后通过行列顺序收集信息以提高信息提取能力，此外还提出了一种局部语义引导解码器来更准确地描绘车道的身份和形状，并使用三个基准测试数据集对其进行了评估，结果表明该网络表现优越。