Structured data question answering (QA), including table QA, Knowledge Graph (KG) QA, and temporal KG QA, is a pivotal research area. Advances in large language models (LLMs) have driven significant progress in unified structural QA frameworks like TrustUQA. However, these frameworks face challenges when applied to small-scale LLMs since small-scale LLMs are prone to errors in generating structured queries. To improve the structured data QA ability of small-scale LLMs, we propose a self-correction distillation (SCD) method. In SCD, an error prompt mechanism (EPM) is designed to detect errors and provide customized error messages during inference, and a two-stage distillation strategy is designed to transfer large-scale LLMs' query-generation and error-correction capabilities to small-scale LLM. Experiments across 5 benchmarks with 3 structured data types demonstrate that our SCD achieves the best performance and superior generalization on small-scale LLM (8B) compared to other distillation methods, and closely approaches the performance of GPT4 on some datasets. Furthermore, large-scale LLMs equipped with EPM surpass the state-of-the-art results on most datasets.

ABSTRACT With the rapid development of deep learning, research on semantic segmentation of remote sensing images has made significant progress. However, there are common problems in remote sensing images, such as large‐scale differences between different types of objects and unbalanced sample numbers, which leads to poor semantic segmentation results, especially for small targets and rare types of objects. To address these challenges, a remote sensing image semantic segmentation method based on multi‐scale contextual information analysis named MSTNet is innovatively proposed. Its core design includes the semantic information enhancement module (SIE) of feature adaptive clustering, which strengthens the feature expression of different categories through adaptive clustering to alleviate sample imbalance; the weighted feature fusion module (WFF) adaptively aggregates cross‐level features, cooperates with the multi‐scale context enhancement module (MSCE), combines convolution and Transformer operations, and deeply mines local and global contexts to cope with scale changes; in addition, the network also contains a pixel space feature optimisation module (SFEM) to enhance spatial details. Experiments on the UAVid, LoveDA, Potsdam and Vaihingen datasets show that MSTNet significantly improves the ability to handle scale changes and imbalance problems and reaches advanced levels in key indicators such as OA, mIoU, and mF1, proving that MSTNet achieves competitive or even better performance.

Zhexin Zhang, Leqi Lei, Lindong Wu, Rui Sun, Yongkang Huang, Chong Long, Xiao Liu, Xuanyu Lei, Jie Tang, Minlie Huang. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Abstract Salient object detection (SOD), one of the most important applications in the field of computer vision, aims to extract the most visually appealing regions of scenes. However, the improvement of the accuracy of existing salient object detection in optical remote sensing images (ORSI‐SOD) is usually accompanied by an increase of network complexity, which affects the application of these models. Motivated by this, a novel lightweight edge‐supervised neural network for ORSI‐SOD is proposed, named CSFFNet. Specifically, the backbone (ResNet34) is first lightened by feature encoding module (FEM), building a lightweight subnet for feature extraction. Then, in the transformer‐based feature pyramid enhancement module (FPEM), the convolutional features obtained in the FEM are enhanced by long‐distance dependence to obtain multi‐scale features containing rich saliency cues. Based on this, the feature fusion module (FFM) is designed to capture cross‐scale long‐range dependencies and effectively fuse high‐level semantic information with low‐level detail information. Thus, the increase in network complexity due to multi‐level decoding is avoided. Finally, the segmentation results are optimized by using salient edges as auxiliary information, which effectively improves the contrast and completeness of the results. Experimental results on two public datasets demonstrate that the lightweight CSFFNet achieves competitive or even better performance compared with state‐of‐the‐art methods.

Conversational recommender system (CRS) enhances the recommender system by acquiring the latest user preference through dialogues, where an agent needs to decide “whether to ask or recommend”, “which attributes to ask”, and “which items to recommend” in each round. To explore these questions, reinforcement learning is adopted in most CRS frameworks. However, existing studies somewhat ignore to consider the connection between the previous rounds and the current round of the conversation, which might lead to the lack of prior knowledge and inaccurate decisions. In this view, we propose to facilitate the connections between different rounds of conversations in a dialogue session through deep transformer-based multi-channel meta-reinforcement learning, so that the CRS agent can decide each action/decision based on previous states, actions, and their rewards. Besides, to better utilize a user’s historical preferences, we propose a more dynamic and personalized graph structure to support the conversation module and the recommendation module. Experiment results on five real-world datasets and an online evaluation with real users in an industrial environment validate the improvement of our method over the state-of-the-art approaches and the effectiveness of our designs.

We address the issue of having a limited number of annotations for stance classification in a new domain, by adapting out-of-domain classifiers with domain adaptation. Existing approaches often align different domains in a single, global feature space (or view), which may fail to fully capture the richness of the languages used for expressing stances, leading to reduced adaptability on stance data. In this paper, we identify two major types of stance expressions that are linguistically distinct, and we propose a tailored dual-view adaptation network (DAN) to adapt these expressions across domains. The proposed model first learns a separate view for domain transfer in each expression channel and then selects the best adapted parts of both views for optimal transfer. We find that the learned view features can be more easily aligned and more stance-discriminative in either or both views, leading to more transferable overall features after combining the views. Results from extensive experiments show that our method can enhance the state-of-the-art single-view methods in matching stance data across different domains, and that it consistently improves those methods on various adaptation tasks.

Allocating resources to customers in the customer service is a difficult problem, because designing an optimal strategy to achieve an optimal trade-off between available resources and customers' satisfaction is non-trivial. In this paper, we formalize the customer routing problem, and propose a novel framework based on deep reinforcement learning (RL) to address this problem. To make it more practical, a demo is provided to show and compare different models, which visualizes all decision process, and in particular, the system shows how the optimal strategy is reached. Besides, our demo system also ships with a variety of models that users can choose based on their needs.

As the rapid development of China's e-commerce in recent years and the underlying evolution of adversarial spamming tactics, more sophisticated spamming activities may carry out in Chinese review websites. Empirical analysis, on recently crawled product reviews from a popular Chinese e-commerce website, reveals the failure of many state-of-the-art spam indicators on detecting collusive spammers. Two novel methods are then proposed: 1) a KNN-based method that considers the pairwise similarity of two reviewers based on their group-level relational information and selects k most similar reviewers for voting; 2) a more general graph-based classification method that jointly classifies a set of reviewers based on their pairwise transaction correlations. Experimental results show that both our methods promisingly outperform the indicator-only classifiers in various settings.

We consider the problem of extracting, in a domain-centric fashion, a given set of attributes from a large number of semi-structured websites. Previous approaches [7, 5] to solve this problem are based on page level inference. We propose a distinct new approach that directly chooses attribute extractors for a site using a scoring mechanism that is designed at the domain level via simple classification methods using a training set from a small number of sites. To keep the number of candidate extractors in each site manageably small we use two observations that hold in most domains: (a) imprecise annotators can be used to identify a small set of candidate extractors for a few attributes (anchors); and (b) non-anchor attributes lie in close proximity to the anchor attributes. Experiments on three domains (Events, Books and Restaurants) show that our approach is very effective in spite of its simplicity.

Fast changing knowledge on the Internet can be acquired more efficiently with the help of automatic document summarization and updating techniques. This paper described a novel approach for multi-document update summarization. The best summary is defined to be the one which has the minimum information distance to the entire document set. The best update summary has the minimum conditional information distance to a document cluster given that a prior document cluster has already been read. Experiments on the DUC 2007 dataset and the TAC 2008 dataset have proved that our method closely correlates with the human summaries and outperforms other programs such as LexRank in many categories under the ROUGE evaluation criterion.

On participatory Websites, users provide opinions about products, with both overall ratings and textual reviews. In this paper, we propose an approach to accurately estimate feature ratings of the products. This approach selects user reviews that extensively discuss specific features of the products (called specialized reviews), using information distance of reviews on the features. Experiments on real data show that overall ratings of the specialized reviews can be used to represent their feature ratings. The average of these overall ratings can be used by recommender systems to provide feature specific recommendations that better help users make purchasing decisions.

If Kolmogorov complexity [25] measures information in one object and Information Distance measures information shared by two objects, how do we measure information shared by many objects? This paper provides an initial pragmatic study of this fundamental data mining question. Firstly, Em(x1,x2,...,xn) is defined to be the minimum amount of thermodynamic energy needed to convert from any xi to any xj. With this definition several theoretical problems have been solved. Second, our newly proposed theory is applied to select a comprehensive review and a specialized review from many reviews: (1) Core feature words, expanded words and dependent words are extracted respectively. (2) Comprehensive and specialized reviews are selected according to the information among them. This method of selecting a single review can be extended to select multiple reviews as well. Finally, experiments show that this comprehensive and specialized review mining method based on our new theory can do the job efficiently.