Pre-trained Language Models (PLMs) may be poisonous with backdoors or bias injected by the suspicious attacker during the fine-tuning process. A core challenge of purifying potentially poisonous PLMs is precisely finding poisonous dimensions. To settle this issue, we propose the Fine-purifying approach, which utilizes the diffusion theory to study the dynamic process of fine-tuning for finding potentially poisonous dimensions. According to the relationship between parameter drifts and Hessians of different dimensions, we can detect poisonous dimensions with abnormal dynamics, purify them by resetting them to clean pre-trained weights, and then fine-tune the purified weights on a small clean dataset. To the best of our knowledge, we are the first to study the dynamics guided by the diffusion theory for safety or defense purposes. Experimental results validate the effectiveness of Fine-purifying even with a small clean dataset.

Recently, Sharpness-Aware Minimization (SAM) algorithm has shown state-of-the-art generalization abilities in vision tasks. It demonstrates that flat minima tend to imply better generalization abilities. However, it has some difficulty implying SAM to some natural language tasks, especially to models with drastic gradient changes, such as RNNs. In this work, we analyze the relation between the flatness of the local minimum and its generalization ability from a novel and straightforward theoretical perspective. We propose that the shift of the training and test distributions can be equivalently seen as a virtual parameter corruption or perturbation, which can explain why flat minima that are robust against parameter corruptions or perturbations have better generalization performances. On its basis, we propose a Gradient-Strength based Adaptive Sharpness-Aware Minimization (GA-SAM) algorithm to help to learn algorithms find flat minima that generalize better. Results in various language benchmarks validate the effectiveness of the proposed GA-SAM algorithm on natural language tasks.

Despite the potential of federated learning, it is known to be vulnerable to backdoor attacks. Many robust federated aggregation methods are proposed to reduce the potential backdoor risk. However, they are mainly validated in the CV field. In this paper, we find that NLP backdoors are hard to defend against than CV, and we provide a theoretical analysis that the malicious update detection error probabilities are determined by the relative backdoor strengths. NLP attacks tend to have small relative backdoor strengths, which may result in the failure of robust federated aggregation methods for NLP attacks. Inspired by the theoretical results, we can choose some dimensions with higher backdoor strengths to settle this issue. We propose a novel federated aggregation algorithm, Dim-Krum, for NLP tasks, and experimental results validate its effectiveness.

Deep Neural Networks (DNNs) are known to be vulnerable to backdoor attacks. In Natural Language Processing (NLP), DNNs are often backdoored during the fine-tuning process of a large-scale Pre-trained Language Model (PLM) with poisoned samples. Although the clean weights of PLMs are readily available, existing methods have ignored this information in defending NLP models against backdoor attacks. In this work, we take the first step to exploit the pre-trained (unfine-tuned) weights to mitigate backdoors in fine-tuned language models. Specifically, we leverage the clean pre-trained weights via two complementary techniques: (1) a two-step Fine-mixing technique, which first mixes the backdoored weights (fine-tuned on poisoned data) with the pre-trained weights, then fine-tunes the mixed weights on a small subset of clean data; (2) an Embedding Purification (E-PUR) technique, which mitigates potential backdoors existing in the word embeddings. We compare Fine-mixing with typical backdoor mitigation methods on three single-sentence sentiment classification tasks and two sentence-pair classification tasks and show that it outperforms the baselines by a considerable margin in all scenarios. We also show that our E-PUR method can benefit existing mitigation methods. Our work establishes a simple but strong baseline defense for secure fine-tuned NLP models against backdoor attacks.

Natural language processing (NLP) models are known to be vulnerable to backdoor attacks, which poses a newly arisen threat to NLP models. Prior online backdoor defense methods for NLP models only focus on the anomalies at either the input or output level, still suffering from fragility to adaptive attacks and high computational cost. In this work, we take the first step to investigate the unconcealment of textual poisoned samples at the intermediate-feature level and propose a feature-based efficient online defense method. Through extensive experiments on existing attacking methods, we find that the poisoned samples are far away from clean samples in the intermediate feature space of a poisoned NLP model. Motivated by this observation, we devise a distance-based anomaly score (DAN) to distinguish poisoned samples from clean samples at the feature level. Experiments on sentiment analysis and offense detection tasks demonstrate the superiority of DAN, as it substantially surpasses existing online defense methods in terms of defending performance and enjoys lower inference costs. Moreover, we show that DAN is also resistant to adaptive attacks based on feature-level regularization. Our code is available at https://github.com/lancopku/DAN.

Previous work has demonstrated the viability of applying deep learning techniques in the financial area. Recently, the task of stock embedding learning has been drawing attention from the research community, which aims to represent the characteristics of stocks with distributed vectors that can be used in various financial analysis scenarios. Existing approaches for learning stock embeddings either require expert knowledge, or mainly focus on the textual part of information corresponding to individual temporal movements. In this paper, we propose to model stock properties as the combination of internal attributes and relational attributes, which takes into consideration both the time-invariant properties of individual stocks and their movement patterns in relation to the market. To learn the two types of attributes from financial news and transaction data, we design several training objectives based on contrastive learning to extract and separate the long-term and temporary information in the data that are able to counter the inherent randomness of the stock market. Experiments and further analyses on portfolio optimization reveal the effectiveness of our method in extracting comprehensive stock information from various data sources.

Recent pretrained language models extend from millions to billions of parameters. Thus the need to fine-tune an extremely large pretrained model with a limited training corpus arises in various downstream tasks. In this paper, we propose a straightforward yet effective fine-tuning technique, Child-Tuning, which updates a subset of parameters (called child network) of large pretrained models via strategically masking out the gradients of the non-child network during the backward process. Experiments on various downstream tasks in GLUE benchmark show that Child-Tuning consistently outperforms the vanilla fine-tuning by 1.5 8.6 average score among four different pretrained models, and surpasses the prior fine-tuning techniques by 0.6 1.3 points. Furthermore, empirical results on domain transfer and task transfer show that Child-Tuning can obtain better generalization performance by large margins.

Recent studies have revealed a security threat to natural language processing (NLP) models, called the Backdoor Attack. Victim models can maintain competitive performance on clean samples while behaving abnormally on samples with a specific trigger word inserted. Previous backdoor attacking methods usually assume that attackers have a certain degree of data knowledge, either the dataset which users would use or proxy datasets for a similar task, for implementing the data poisoning procedure. However, in this paper, we find that it is possible to hack the model in a data-free way by modifying one single word embedding vector, with almost no accuracy sacrificed on clean samples. Experimental results on sentiment analysis and sentence-pair classification tasks show that our method is more efficient and stealthier. We hope this work can raise the awareness of such a critical security risk hidden in the embedding layers of NLP models. Our code is available at https://github.com/lancopku/Embedding-Poisoning.

Side effects during neural network tuning are typically measured by overall accuracy changes. However, we find that even with similar overall accuracy, existing tuning methods result in non-negligible instance-wise side effects. Motivated by neuroscientific evidence and theoretical results, we demonstrate that side effects can be controlled by the number of changed parameters and thus, we propose to conduct neural network surgery by only modifying a limited number of parameters. Neural network surgery can be realized using diverse techniques and we investigate three lines of methods. Experimental results on representative tuning problems validate the effectiveness of the surgery approach. The dynamic selecting method achieves the best overall performance that not only satisfies the tuning goal but also induces fewer instance-wise side effects by changing only 10^-5 of the parameters.

Conventional knowledge graph embedding (KGE) often suffers from limited knowledge representation, leading to performance degradation especially on the low-resource problem. To remedy this, we propose to enrich knowledge representation via pretrained language models by leveraging world knowledge from pretrained models. Specifically, we present a universal training framework named Pretrain-KGE consisting of three phases: semantic-based fine-tuning phase, knowledge extracting phase and KGE training phase. Extensive experiments show that our proposed Pretrain-KGE can improve results over KGE models, especially on solving the low-resource problem.

Skip connection is a widely-used technique to improve the performance and the convergence of deep neural networks, which is believed to relieve the difficulty in optimization due to non-linearity by propagating a linear component through the neural network layers. However, from another point of view, it can also be seen as a modulating mechanism between the input and the output, with the input scaled by a pre-defined value one. In this work, we investigate how the scale factors in the effectiveness of the skip connection and reveal that a trivial adjustment of the scale will lead to spurious gradient exploding or vanishing in line with the deepness of the models, which could by addressed by normalization, in particular, layer normalization, which induces consistent improvements over the plain skip connection. Inspired by the findings, we further propose to adaptively adjust the scale of the input by recursively applying skip connection with layer normalization, which promotes the performance substantially and generalizes well across diverse tasks including both machine translation and image classification datasets.