Large language models (LLMs) show promising performance on small-scale graph reasoning tasks but fail when handling real-world graphs with complex queries. This phenomenon arises from LLMs' working memory constraints, which result in their inability to retain long-range graph topology over extended contexts while sustaining coherent multi-step reasoning. However, real-world graphs are often structurally complex, such as Web, Transportation, Social, and Citation networks. To address these limitations, we propose GraphCogent, a collaborative agent framework inspired by human Working Memory Model that decomposes graph reasoning into specialized cognitive processes: sense, buffer, and execute. The framework consists of three modules: Sensory Module standardizes diverse graph text representations via subgraph sampling, Buffer Module integrates and indexes graph data across multiple formats, and Execution Module combines tool calling and tool creation for efficient reasoning. We also introduce Graph4real, a comprehensive benchmark that contains four domains of real-world graphs (Web, Transportation, Social, and Citation) to evaluate LLMs' graph reasoning capabilities. Our Graph4real covers 21 different graph reasoning tasks, categorized into three types (Structural Querying, Algorithmic Reasoning, and Predictive Modeling tasks), with graph scales up to 10 times larger than existing benchmarks. Experiments show that Llama3.1-8B based GraphCogent achieves a 50% improvement over massive-scale LLMs like DeepSeek-R1 (671B). Compared to state-of-the-art code-based baseline, our framework outperforms by 20% in accuracy while reducing token usage by 80% for in-toolset tasks and 30% for out-toolset tasks.

Session-based recommendation (SBR) predicts the next interaction of anonymous users based on their session data. In practical scenarios, low-exposure tail items constitute the majority of interactions, leading to the biased recommendation whereby a small set of high-exposure head items is preferentially recommended. Existing solutions focus on promoting those tail items but with a sacrifice in recommendation accuracy. We attribute such a tradeoff to session-irrelevant noise within the tail items, which existing long-tail approaches fail to identify and constrain effectively. To resolve this fundamental conflict, we propose Hybrid Intent-based Dual Constraint framework (HID) , a plug-and-play framework that converts the traditional tradeoff between long-tail performance and accuracy into a synergistic relationship by introducing hybrid intent-based dual constraints. HID incorporates two key innovations: (i) Hybrid Intent Learning , which reformulates intent extraction through attribute-aware spectral clustering and discriminates session-irrelevant noise by assigning target and noise intents to each session and (ii) Intent Constraint Loss , which introduces two constraints—diversity and accuracy—to regulate item and session representation learning. These two objectives are unified into a single training loss through rigorous theoretical derivation. To handle ambiguous user intent in the real-world, we also propose a slack version of HID. Extensive experiments show that HID enhances both long-tail performance and accuracy, establishing new state-of-the-art performance in long-tail recommender systems. The implementation code is available at: https://github.com/jarviswww/Code4TradeSynergy .

Vision Transformers (ViTs) and their variants have achieved remarkable success across a broad spectrum of computer vision tasks. However, their high computational cost and significant data requirements present challenges for deployment in resource-constrained environments. Current pruning methods for ViTs predominantly focus on reducing token counts, which often disrupt the inherent spatial structure of ViTs, hindering their adaptability to hardware. Furthermore, how to compress ViTs efficiently in few-shot scenarios remains an open question. Hence, we introduce a fine-grained block pruning framework for ViTs, named FBP-ViT. Unlike traditional block pruning techniques that indiscriminately remove entire blocks, FBP-ViT selectively eliminates Multi-Head Self-Attention (MSA) or Multi-Layer Perceptron (MLP) blocks, offering enhanced flexibility and efficiency. We unify pruning and finetuning, ensuring practicality in resource-constrained environments for real-world applications. We evaluate the proposed FBP-ViT across ViTs of varying sizes and architectures, demonstrating its effectiveness in improving computational efficiency while maintaining high performance. Specifically, with a speedup factor of 1.34, FBP-ViT preserves 80.39% top-1 accuracy on ImageNet-1k using DeiT-Base, achieving more precise and efficient pruning with tiny sets.

In recent years, thanks to the vigorous development of deep learning, single image super-resolution has advanced greatly. Most super-resolution (SR) methods use convolution layers to construct the network, which achieves superior results over the traditional methods based on manual features. However, most methods based on convolutional neural networks (CNN) blindly deepen the depth of the network leading to a large number of model parameters, which inevitably brings huge computing overhead and memory consumption, and greatly limits the application in resource-limited devices. In order to alleviate this problem, a knowledge distillation framework based on contrastive learning is proposed to compress and accelerate the SR model with enormous parameters. The student network is directly constructed by reducing the number of layers of the teacher network. In particular, the proposed method distills the statistical information of the intermediate feature maps from the teacher network to train the lightweight student network. In addition, through explicit knowledge transfer, a novel contrastive loss is introduced to improve the reconstruction performance of the student network. Experiments show that the proposed contrastive distillation framework can effectively compress the model scale with an acceptable loss of performance.

Recently, many knowledge graph embedding models for knowledge graph completion have been proposed, ranging from the initial translation-based models such as TransE to recent convolutional neural network (CNN) models such as ConvE. However, these models only focus on semantic information of knowledge graph and neglect the natural graph structure information. Although graph convolutional network (GCN)-based models for knowledge graph embedding have been introduced to address this issue, they still suffer from fact incompleteness, resulting in the unconnectedness of knowledge graph. To solve this problem, we propose a novel model called deep relational graph infomax (DRGI) with mutual information (MI) maximization which takes the benefit of complete structure information and semantic information together. Specifically, the proposed DRGI consists of two encoders which are two identical adaptive relational graph attention networks (ARGATs), corresponding to catching semantic information and complete structure information respectively. Our method establishes new state-of-the-art on the standard datasets for knowledge graph completion. In addition, by exploring the complete structure information, DRGI embraces the merits of faster convergence speed over existing methods and better predictive performance for entities with small indegree.

Recently, many knowledge graph embedding models for knowledge graph completion have been proposed, ranging from the initial translation-based models such as TransE to recent convolutional neural network (CNN) models such as ConvE. However, these models only focus on semantic information of knowledge graph and neglect the natural graph structure information. Although graph convolutional network (GCN)-based models for knowledge graph embedding have been introduced to address this issue, they still suffer from fact incompleteness, resulting in the unconnectedness of knowledge graph. To solve this problem, we propose a novel model called deep relational graph infomax (DRGI) with mutual information (MI) maximization which takes the benefit of complete structure information and semantic information together. Specifically, the proposed DRGI consists of two encoders which are two identical adaptive relational graph attention networks (ARGATs), corresponding to catching semantic information and complete structure information respectively. Our method establishes new state-of-the-art on the standard datasets for knowledge graph completion.