Yancheng He, Shilong Li, Jiaheng Liu, Yingshui Tan, Weixun Wang, Hui Huang, Xingyuan Bu, Hangyu Guo, Chengwei Hu, Boren Zheng, Zhuoran Lin, Dekai Sun, Zhicheng Zheng, Wenbo Su, Bo Zheng. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Many previous models of named entity recognition (NER) suffer from the problem of Out-of-Entity (OOE), i.e., the tokens in the entity mentions of the test samples have not appeared in the training samples, which hinders the achievement of satisfactory performance. To improve OOE-NER performance, in this paper, we propose a new framework, namely S+NER, which fully leverages sentence-level information. Our S+NER achieves better OOE-NER performance mainly due to the following two particular designs. 1) It first exploits the pre-trained language model’s capability of understanding the target entity’s sentence-level context with a template set. 2) Then, it refines the sentence-level representation based on the positive and negative templates, through a contrastive learning strategy and template pooling method, to obtain better NER results. Our extensive experiments on five benchmark datasets have demonstrated that, our S+NER outperforms some state-of-the-art OOE-NER models.

With the development of large language models, their ability to follow simple instructions has significantly improved.However, adhering to complex instructions remains a major challenge.Current approaches to generating complex instructions are often irrelevant to the current instruction requirements or suffer from limited scalability and diversity.Moreover, methods such as back-translation, while effective for simple instruction generation, fail to leverage the rich knowledge and formatting in human written documents.In this paper, we propose a novel Automatic Iterative Refinement (AIR) framework to generate complex instructions with constraints, which not only better reflects the requirements of real scenarios but also significantly enhances LLMs' ability to follow complex instructions.The AIR framework consists of two stages: 1) Generate an initial instruction from a document; 2) Iteratively refine instructions with LLM-as-judge guidance by comparing the model's output with the document to incorporate valuable constraints.Finally, we construct the AIR-10K dataset with 10K complex instructions and demonstrate that instructions generated with our approach significantly improve the model's ability to follow complex instructions, outperforming existing methods for instruction generation 1 . Model Answer Refined D Model Answer Check Constraints Identify ConstraintsI: Write a casual review of a water-proof camera.

With the increasing use of Large Language Models (LLMs) in fields such as e-commerce, domain-specific concept evaluation benchmarks are crucial for assessing their domain capabilities. Existing LLMs may generate factually incorrect information within the complex e-commerce applications. Therefore, it is necessary to build an e-commerce concept benchmark. Existing benchmarks encounter two primary challenges: (1) handle the heterogeneous and diverse nature of tasks(2) distinguish between generality and specificity within the e-commerce field. To address these problems, we propose ChineseEcomQA, a scalable question-answering benchmark focused on fundamental e-commerce concepts. ChineseEcomQA is built on three core characteristics: Focus on Fundamental Concept, E-commerce Generality and E-commerce Expertise. Fundamental concepts are designed to be applicable across a diverse array of e-commerce tasks, thus addressing the challenge of heterogeneity and diversity. Additionally, by carefully balancing generality and specificity, ChineseEcomQA effectively differentiates between broad e-commerce concepts, allowing for precise validation of domain capabilities. We achieve this through a scalable benchmark construction process that combines LLM validation, Retrieval-Augmented Generation (RAG) validation, and rigorous manual annotation. Based on ChineseEcomQA, we conduct extensive evaluations on mainstream LLMs and provide some valuable insights. We hope that ChineseEcomQA could guide future domain-specific evaluations, and facilitate broader LLM adoption in e-commerce applications.

Abstract Recently, unsupervised image‐to‐image (i2i) translation methods based on contrastive learning have achieved state‐of‐the‐art results. However, in previous works, the negatives are sampled from the input image itself, which inspires us to design a data augmentation method to improve the quality of the selected negatives. Moreover, the previous methods only preserve the content consistency via patch‐wise contrastive learning, which ignores the domain consistency between the generated images and the real images of the target domain. This paper proposes a novel unsupervised i2i translation framework based on multi‐cropping contrastive learning and domain consistency, called MCDUT. Specifically, the multi‐cropping views are obtained with the aim of further generating high‐quality negative examples. To constrain the embeddings in the deep feature space, a new domain consistency loss is formulated, which encourages the generated images to be close to the real images. In many i2i translation tasks, this method achieves state‐of‐the‐art results, and the advantages of this method have been proven through extensive comparison experiments and ablation research. The code of MCDUT is available at https://github.com/zhihefang/MCDUT .

Underwater images are subject to intricate and diverse degradation, inevitably affecting the effectiveness of underwater visual tasks. However, most approaches primarily operate in the raw pixel space of images, which limits the exploration of the frequency characteristics of underwater images, leading to an inadequate utilization of deep models' representational capabilities in producing high-quality images. In this paper, we introduce a novel Underwater Image Enhancement (UIE) framework, named WF-Diff, designed to fully leverage the character-istics of frequency domain information and diffusion models. WF-Diff consists of two detachable networks: Wavelet-based Fourier information interaction network (WFI2-net) and Frequency Residual Diffusion Adjustment Module (FR-DAM). With our full exploration of the frequency domain in-formation, WFI2-net aims to achieve preliminary enhancement of frequency information in the wavelet space. Our proposed FRDAM can further refine the highand low-frequency information of the initial enhanced images, which can be viewed as a plug-and-play universal module to adjust the detail of the underwater images. With the above techniques, our algorithm can show SOTA performance on real-world underwater image datasets, and achieves competitive performance in visual quality. The code is available at https://github.com/zhihefang/WF-Diff.

Image deblurring presents a complex challenge intending to renew visual clarity in images affected by camera shake or object motion. However, traditional deblurring methodologies tend to emphasize local features, ignoring critical contextual information, which consequently limits their efficacy in addressing common blurry image issues. This paper proposes Efficient Cross Transformer (ECT) to overcome the limitations of inadequate global features in image restoration across different scales. ECT designs the cross-attention layers to achieve interaction between input tokens and network tokens, aiming to efficiently capture image details in the input space and feature information in the latent space of each layer. The cross-attention mechanism works in latent layers, integrating image features from different scales while reducing the computational burden compared to traditional Transformers. Furthermore, ECT employs windowing techniques in a strategic manner to capture local hazy components, duly amplifying its image restoration abilities. Empirical evidence shows that ECT has achieved state-of-the-art results in deblurring images, demonstrating excellent performance without requiring a large training dataset. Consequently, ECT emerges as a promising solution for rectifying blurred images across artificial and real-world environments.

Continual relation extraction (CRE) aims to extract relations towards the continuous and iterative arrival of new data, of which the major challenge is the catastrophic forgetting of old tasks. In order to alleviate this critical problem for enhanced CRE performance, we propose a novel Continual Relation Extraction framework with Contrastive Learning, namely CRECL, which is built with a classification network and a prototypical contrastive network to achieve the incremental-class learning of CRE. Specifically, in the contrastive network a given instance is contrasted with the prototype of each candidate relations stored in the memory module. Such contrastive learning scheme ensures the data distributions of all tasks more distinguishable, so as to alleviate the catastrophic forgetting further. Our experiment results not only demonstrate our CRECL’s advantage over the state-of-the-art baselines on two public datasets, but also verify the effectiveness of CRECL’s contrastive learning on improving performance.

Li Cui, Deqing Yang, Jiaxin Yu, Chengwei Hu, Jiayang Cheng, Jingjie Yi, Yanghua Xiao. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.