Jian Wang

Dongding Lin, Jian Wang, Wenjie Li

Conversational recommender systems (CRS) aim to employ natural language conversations to suggest suitable products to users. Understanding user preferences for prospective items and learning efficient item representations are crucial for CRS. Despite various attempts, earlier studies mostly learned item representations based on individual conversations, ignoring item popularity embodied among all others. Besides, they still need support in efficiently capturing user preferences since the information reflected in a single conversation is limited. Inspired by collaborative filtering, we propose a collaborative augmentation (COLA) method to simultaneously improve both item representation learning and user preference modeling to address these issues. We construct an interactive user-item graph from all conversations, which augments item representations with user-aware information, i.e., item popularity. To improve user preference modeling, we retrieve similar conversations from the training corpus, where the involved items and attributes that reflect the user's potential interests are used to augment the user representation through gate control. Extensive experiments on two benchmark datasets demonstrate the effectiveness of our method. Our code and data are available at https://github.com/DongdingLin/COLA.

Zhehao Tan, Yihan Jiao, Dan Yang et al.

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by integrating external knowledge, where the LLM's ability to generate responses based on the combination of a given query and retrieved documents is crucial. However, most benchmarks focus on overall RAG system performance, rarely assessing LLM-specific capabilities. Current benchmarks emphasize broad aspects such as noise robustness, but lack a systematic and granular evaluation framework on document utilization. To this end, we introduce \textit{Placeholder-RAG-Benchmark}, a multi-level fine-grained benchmark, emphasizing the following progressive dimensions: (1) multi-level filtering abilities, (2) combination abilities, and (3) reference reasoning. To provide a more nuanced understanding of LLMs' roles in RAG systems, we formulate an innovative placeholder-based approach to decouple the contributions of the LLM's parametric knowledge and the external knowledge. Experiments demonstrate the limitations of representative LLMs in the RAG system's generation capabilities, particularly in error resilience and context faithfulness. Our benchmark provides a reproducible framework for developing more reliable and efficient RAG systems. Our code is available in https://github.com/Alipay-Med/PRGB.

Jian Wang, Xiaofei Xie, Qiang Hu et al.

Code Large Language Models (Code LLMs) have opened a new era in programming with their impressive capabilities. However, recent research has revealed critical limitations in their ability to reason about runtime behavior and understand the actual functionality of programs, which poses significant challenges for their post-training and practical deployment. Specifically, Code LLMs encounter two principal issues: (1) a lack of proficiency in reasoning about program execution behavior, as they struggle to interpret what programs actually do during runtime, and (2) the inconsistent and fragmented representation of semantic information, such as execution traces, across existing methods, which hinders their ability to generalize and reason effectively. These challenges underscore the necessity for more systematic approaches to enhance the reasoning capabilities of Code LLMs. To address these issues, we introduce a generic framework to support integrating semantic information~(e.g., execution trace) to code task-relevant prompts, and conduct a comprehensive study to explore the role of semantic information in enhancing the reasoning ability of Code LLMs accordingly. Specifically, we focus on investigating the usefulness of trace-based semantic information in boosting supervised fine-tuning~(SFT) and post-phase inference of Code LLMs. The experimental results surprisingly disagree with previous works and demonstrate that semantic information has limited usefulness for SFT and test time scaling of Code LLM.

Wei-Ting Chen, Yu-Jiet Vong, Yi-Tsung Lee et al.

Video Quality Assessment (VQA) aims to evaluate video quality based on perceptual distortions and human preferences. Despite the promising performance of existing methods using Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), they often struggle to align closely with human perceptions, particularly in diverse real-world scenarios. This challenge is exacerbated by the limited scale and diversity of available datasets. To address this limitation, we introduce a novel VQA framework, DiffVQA, which harnesses the robust generalization capabilities of diffusion models pre-trained on extensive datasets. Our framework adapts these models to reconstruct identical input frames through a control module. The adapted diffusion model is then used to extract semantic and distortion features from a resizing branch and a cropping branch, respectively. To enhance the model's ability to handle long-term temporal dynamics, a parallel Mamba module is introduced, which extracts temporal coherence augmented features that are merged with the diffusion features to predict the final score. Experiments across multiple datasets demonstrate DiffVQA's superior performance on intra-dataset evaluations and its exceptional generalization across datasets. These results confirm that leveraging a diffusion model as a feature extractor can offer enhanced VQA performance compared to CNN and ViT backbones.

Na zhao, Zhen Long, Zhi-Dan Zhao et al.

Due to a large amount of information, it is difficult for users to find what they are interested in among the many choices. In order to improve users' experience, recommendation systems have been widely used in music recommendations, movie recommendations, online shopping, and other scenarios. Recently, Knowledge Graph (KG) has been proven to be an effective tool to improve the performance of recommendation systems. However, a huge challenge in applying knowledge graphs for recommendation is how to use knowledge graphs to obtain better user codes and item codes. In response to this problem, this research proposes a user Recurrent Neural Network (RNN) encoder and item encoder recommendation algorithm based on Knowledge Graph (URIR). This study encodes items by capturing high-level neighbor information to generate items' representation vectors and applies an RNN and items' representation vectors to encode users to generate users' representation vectors, and then perform inner product operation on users' representation vectors and items' representation vectors to get probabilities of users interaction with items. Numerical experiments on three real-world datasets demonstrate that URIR is superior performance to state-of-the-art algorithms in indicators such as AUC, Precision, Recall, and MRR. This implies that URIR can effectively use knowledge graph to obtain better user codes and item codes, thereby obtaining better recommendation results.

Jiabin Liu, Zheng Wei, Zhengpin Li et al.

Recently, textual information has been proved to play a positive role in recommendation systems. However, most of the existing methods only focus on representation learning of textual information in ratings, while potential selection bias induced by the textual information is ignored. In this work, we propose a novel and general self-adaptive module, the Self-adaptive Attention Module (SAM), which adjusts the selection bias by capturing contextual information based on its representation. This module can be embedded into recommendation systems that contain learning components of contextual information. Experimental results on three real-world datasets demonstrate the effectiveness of our proposal, and the state-of-the-art models with SAM significantly outperform the original ones.

Yutao Ma, Xiao Geng, Jian Wang et al.

Recent years have witnessed the rapid development of service-oriented computing technologies. The boom of Web services increases software developers' selection burden in developing new service-based systems such as mashups. Timely recommending appropriate component services for developers to build new mashups has become a fundamental problem in service-oriented software engineering. Existing service recommendation approaches are mainly designed for mashup development in the single-round scenario. It is hard for them to effectively update recommendation results according to developers' requirements and behaviours (e.g. instant service selection). To address this issue, the authors propose a service bundle recommendation framework based on deep learning, DLISR, which aims to capture the interactions among the target mashup to build, selected (component) services, and the following service to recommend. Moreover, an attention mechanism is employed in DLISR to weigh selected services when recommending a candidate service. The authors also design two separate models for learning interactions from the perspectives of content and invocation history, respectively, and a hybrid model called HISR. Experiments on a real-world dataset indicate that HISR can outperform several state-of-the-art service recommendation methods to develop new mashups iteratively.

Xiaorui Wu, Hong Xu, Honglin Zhang et al.

Production recommendation systems rely on embedding methods to represent various features. An impeding challenge in practice is that the large embedding matrix incurs substantial memory footprint in serving as the number of features grows over time. We propose a similarity-aware embedding matrix compression method called Saec to address this challenge. Saec clusters similar features within a field to reduce the embedding matrix size. Saec also adopts a fast clustering optimization based on feature frequency to drastically improve clustering time. We implement and evaluate Saec on Numerous, the production distributed machine learning system in Tencent, with 10-day worth of feature data from QQ mobile browser. Testbed experiments show that Saec reduces the number of embedding vectors by two orders of magnitude, compresses the embedding size by ~27x, and delivers the same AUC and log loss performance.

Jian Wang, Krishnaram Kenthapadi, Kaushik Rangadurai et al.

We address the following problem: How do we incorporate user item interaction signals as part of the relevance model in a large-scale personalized recommendation system such that, (1) the ability to interpret the model and explain recommendations is retained, and (2) the existing infrastructure designed for the (user profile) content-based model can be leveraged? We propose Dionysius, a hierarchical graphical model based framework and system for incorporating user interactions into recommender systems, with minimal change to the underlying infrastructure. We learn a hidden fields vector for each user by considering the hierarchy of interaction signals, and replace the user profile-based vector with this learned vector, thereby not expanding the feature space at all. Thus, our framework allows the use of existing recommendation infrastructure that supports content based features. We implemented and deployed this system as part of the recommendation platform at LinkedIn for more than one year. We validated the efficacy of our approach through extensive offline experiments with different model choices, as well as online A/B testing experiments. Our deployment of this system as part of the job recommendation engine resulted in significant improvement in the quality of retrieved results, thereby generating improved user experience and positive impact for millions of users.