Exclusion and Inclusion -- A model agnostic approach to feature importance in DNNs

MASE：通过模型无关显著性估计实现可解释的NLP模型

MASE: Interpretable NLP Models via Model-Agnostic Saliency Estimation

MASE通过模型无关显著性估计方法，为NLP文本模型提供可解释的局部解释，提高了模型决策过程的透明度。

DLBacktrace：一种适用于任何深度学习模型的模型无关可解释性方法

DLBacktrace: A Model Agnostic Explainability for any Deep Learning Models

本文提出模型无关的可解释性方法DLBacktrace，适用于MLP、CNN和Transformer等架构，并在与SHAP、LIME、GradCAM的对比中验证其能有效提升对模型行为的理解。

Enhancing Model Robustness By Incorporating Adversarial Knowledge Into Semantic Representation

Despite that deep neural networks (DNNs) have achieved enormous success in many domains like natural language processing (NLP), they have also been proven to be vulnerable to maliciously generated adversarial examples. Such inherent vulnerability has threatened various real-world deployed DNNs-based applications. To strength the model robustness, several countermeasures have been proposed in the English NLP domain and obtained satisfactory performance. However, due to the unique language properties of Chinese, it is not trivial to extend existing defenses to the Chinese domain. Therefore, we p…

OTOv3: 从结构化剪枝到算子擦除的自动架构无关神经网络训练与压缩

OTOv3: Automatic Architecture-Agnostic Neural Network Training and Compression from Structured Pruning to Erasing Operators

针对现有DNN压缩方法流程复杂、需大量专业知识的问题，提出OTOv3框架，通过依赖图分析、DHSPG/H2SPG算法等实现免微调的自动训练压缩，性能达SOTA。

A Good Sample is Hard to Find: Noise Injection Sampling and Self-Training for Neural Language Generation Models

Deep neural networks (DNN) are quickly becoming the de facto standard modeling method for many natural language generation (NLG) tasks. In order for such models to truly be useful, they must be capable of correctly generating utterances for novel meaning representations (MRs) at test time. In practice, even sophisticated DNNs with various forms of semantic control frequently fail to generate utterances faithful to the input MR. In this paper, we propose an architecture agnostic self-training method to sample novel MR/text utterance pairs to augment the original training data. Remarkably, after…