Cross-prompt automated essay scoring (AES) aims to train models using essays from different source prompts and test them on new target prompt essays. A core challenge of the task is to learn as much shared knowledge as possible between essays from different prompts in order to better represent new prompt essays. Previous studies primarily focus on learning this knowledge on a general, coarse-grained level, ignoring that the shared knowledge among prompts is highly detailed and contains a more comprehensive range of information that is not fully investigated. In this paper, we propose a novel multi-aspect knowledge finding and aligning optimization strategy to better acquire this detailed various shared knowledge. We also introduce LLM to extract explicit, interpretable knowledge from implicit, multi-aspect shared knowledge and use this knowledge to improve the representation and evaluation performance of new prompt essays. We conduct extensive experiments on public datasets. The results show that our approach outperforms current state-of-the-art models and is effective on cross-prompt AES.

Most current automated essay quality assessment systems treat score prediction and feedback generation as separate tasks, overlooking the fact that scores provide a quantitative evaluation of quality, while feedback offers a qualitative assessment.Both aspects reflect essay quality from different perspectives, and they are inherently consistent and can reinforce each other.In this paper, we propose a novel bidirectional reinforcement learning framework that effectively utilizes this consistency constraint to jointly optimize score prediction and feedback generation, ensuring mutual reinforcement and alignment between them.In this way, our model is hope to obtain a simultaneous accurate ratings and consistent text feedback.We conducted extensive experiments on publicly available datasets.The results demonstrate that our approach surpasses the current state-of-theart models, enhancing both scoring accuracy and feedback quality.

Most of the existing aspect-based sentiment analysis (ABSA) models only predict the sentiment polarity of a single aspect at a time, focusing primarily on enhancing the representation of this single aspect based on the other contexts or aspects. This one-to-one paradigm ignores the fact that multi-aspect, multi-sentiment sentences contain not only distinct specific descriptions for distinct specific aspects, but also shared global context information for multiple aspects. To fully consider these issues, we propose a one-to-many ABSA framework, called You Only Read Once (YORO), that can simultaneously model representations of all aspects based on their specific descriptions and better fuse their relationships using globally shared contextual information in the sentence. Predicting the sentiment polarity of multiple aspects simultaneously is beneficial to improving the efficacy of calculation and prediction. Extensive experiments are conducted on three public datasets (MAMS, Rest14, and Lap14). Experimental results demonstrate the effectiveness of YORO in handling multi-aspect, multi-sentiment scenarios and highlight the promise of one-to-many ABSA in balancing efficiency and accuracy.

Software vulnerability detection aims to identify latent weaknesses in source code to protect software systems from attacks. Most of the current neural network-based approaches rely on preprocessing source codes using code parsers to obtain program dependency graphs for downstream vulnerability detection tasks. However, as source codes may contain incorrect format and special functions, this paradigm may cause parsing errors due to their failure to satisfy the parser's presets or requirements. In addition, modern software development often involves the integration of multiple programming languages, presenting challenges for conventional code parsers that are designed for specific programming languages. To address these issues, we present a novel detection framework that can identify software vulnerabilities in a programming language-agnostic manner without relying on code parsers. We conduct extensive experiments on the CodeXGlue,Reveal, and FUNDED public datasets and evaluate them against leading vulnerability detection benchmarks. Experimental results show that BABEL sets new performance standards, demonstrating superior accuracy and F1 scores across various tasks, including those in language-independent settings, demonstrating the applicability of BABEL in real-world application scenarios. Our code is available at https://github.com/gdufsnlp/BABEL

Global stock dynamics deeply influence stock price movements, which are often reflected in the consistent trends of the market index and individual stock prices. Nevertheless, external events (e.g., economic policies and geopolitics) and stock industry features could make special cases where local stock price movements deviate from market index trends. In these cases, investors and financial analysts can often identify and make reasonable investment decisions based on their investment experience and knowledge of the global market index trend and specific stocks. We argue that these investment insights can be incorporated into the model to make a more accurate decision. However, few existing works investigate this issue. In this paper, we design a novel prompt to introduce investment experience knowledge and introduce a trend-consistent contrastive learning strategy into the model to improve awareness of the relationship between the stock and the market. Specifically, by pulling stocks with consistent trends closer together and moving stocks with inconsistent trends apart, the model is expected to learn more personalized relationships between individual stocks and the global market. In this way, our model is able to provide more accurate and comprehensive stock analysis and prediction. We conduct extensive experiments on the public dataset, and the results demonstrate the effectiveness of our model.

Data augmentation aims at expanding training data with clean text using noising schemes to improve the performance of grammatical error correction (GEC).In practice, there are a great number of real error patterns in the manually annotated training data.We argue that these real error patterns can be introduced into clean text to effectively generate more real and high quality synthetic data, which is not fully explored by previous studies.Moreover, we also find that linguistic knowledge can be incorporated into data augmentation for generating more representative and more diverse synthetic data.In this paper, we propose a novel data augmentation method that fully considers the real error patterns and the linguistic knowledge for the GEC task.We conduct extensive experiments on public data sets and the experimental results show that our method outperforms several strong baselines with far less external unlabeled clean text data, highlighting its extraordinary effectiveness in the GEC task that lacks large-scale labeled training data.

In recent years, many data mining practitioners have treated deep neural networks (DNNs) as a standard recipe of creating the state-of-the-art solutions. As a result, models like Support Vector Machines (SVMs) have been overlooked. While the results from DNNs are encouraging, DNNs also come with their huge number of parameters in the model and overheads in long training/inference time. SVMs have excellent properties such as convexity, good generality and efficiency. In this paper, we propose techniques to enhance SVMs with an automatic pipeline which exploits the context of the learning problem. The pipeline consists of several components including data aware subproblem construction, feature customization, data balancing among subproblems with augmentation, and kernel hyper-parameter tuner. Comprehensive experiments show that our proposed solution is more efficient, while producing better results than the other SVM based approaches. Additionally, we conduct a case study of our proposed solution on a popular sentiment analysis problem---the aspect term sentiment analysis (ATSA) task. The study shows that our SVM based solution can achieve competitive predictive accuracy to DNN (and even majority of the BERT) based approaches. Furthermore, our solution is about 40 times faster in inference and has 100 times fewer parameters than the models using BERT. Our findings can encourage more research work on conventional machine learning techniques which may be a good alternative for smaller model size and faster training/inference.

Multimodal sentiment analysis aims to predict sentiment of language text with the help of other modalities, such as vision and acoustic features.Previous studies focused on learning the joint representation of multiple modalities, ignoring some useful knowledge contained in language modal.In this paper, we try to incorporate sentimental words knowledge into the fusion network to guide the learning of joint representation of multimodal features.Our method consists of two components: shallow fusion part and aggregation part.For the shallow fusion part, we use crossmodal coattention mechanism to obtain bidirectional context information of each two modals to get the fused shallow representations.For the aggregation part, we design a multitask of sentimental words classification to help and guide the deep fusion of the three modalities and obtain the final sentimental words aware fusion representation.We carry out several experiments on CMU-MOSI, CMU-MOSEI and YouTube datasets.The experimental results show that introducing sentimental words prediction as a multitask can really improve the fusion representation of multiple modalities.