Yuhang Ye

Libin Qiu, Zhirong Gao, Junfu Chen et al.

Large language model agents often fail to accumulate knowledge from experience, treating each task as an independent challenge. Recent methods extract experience as flattened textual knowledge, which cannot capture procedural logic of complex subtasks. They also lack maintenance mechanisms, causing repository degradation as experience accumulates. We introduce AutoRefine, a framework that extracts and maintains dual-form Experience Patterns from agent execution histories. For procedural subtasks, we extract specialized subagents with independent reasoning and memory. For static knowledge, we extract skill patterns as guidelines or code snippets. A continuous maintenance mechanism scores, prunes, and merges patterns to prevent repository degradation. Evaluated on ALFWorld, ScienceWorld, and TravelPlanner, AutoRefine achieves 98.4%, 70.4%, and 27.1% respectively, with 20-73% step reductions. On TravelPlanner, automatic extraction exceeds manually designed systems (27.1% vs 12.1%), demonstrating its ability to capture procedural coordination.

Lu Yi, Jie Peng, Yanping Zheng et al.

Future link prediction is a fundamental challenge in various real-world dynamic systems. To address this, numerous temporal graph neural networks (temporal GNNs) and benchmark datasets have been developed. However, these datasets often feature excessive repeated edges and lack complex sequential dynamics, a key characteristic inherent in many real-world applications such as recommender systems and ``Who-To-Follow'' on social networks. This oversight has led existing methods to inadvertently downplay the importance of learning sequential dynamics, focusing primarily on predicting repeated edges. In this study, we demonstrate that existing methods, such as GraphMixer and DyGFormer, are inherently incapable of learning simple sequential dynamics, such as ``a user who has followed OpenAI and Anthropic is more likely to follow AI at Meta next.'' Motivated by this issue, we introduce the Temporal Graph Benchmark with Sequential Dynamics (TGB-Seq), a new benchmark carefully curated to minimize repeated edges, challenging models to learn sequential dynamics and generalize to unseen edges. TGB-Seq comprises large real-world datasets spanning diverse domains, including e-commerce interactions, movie ratings, business reviews, social networks, citation networks and web link networks. Benchmarking experiments reveal that current methods usually suffer significant performance degradation and incur substantial training costs on TGB-Seq, posing new challenges and opportunities for future research. TGB-Seq datasets, leaderboards, and example codes are available at https://tgb-seq.github.io/.

Jie Peng, Zhewei Wei, Yuhang Ye

Due to the proficiency of self-attention mechanisms (SAMs) in capturing dependencies in sequence modeling, several existing dynamic graph neural networks (DGNNs) utilize Transformer architectures with various encoding designs to capture sequential evolutions of dynamic graphs. However, the effectiveness and efficiency of these Transformer-based DGNNs vary significantly, highlighting the importance of properly defining the SAM on dynamic graphs and comprehensively encoding temporal and interactive dynamics without extra complex modules. In this work, we propose TIDFormer, a dynamic graph TransFormer that fully exploits Temporal and Interactive Dynamics in an efficient manner. We clarify and verify the interpretability of our proposed SAM, addressing the open problem of its uninterpretable definitions on dynamic graphs in previous works. To model the temporal and interactive dynamics, respectively, we utilize the calendar-based time partitioning information and extract informative interaction embeddings for both bipartite and non-bipartite graphs using merely the sampled first-order neighbors. In addition, we jointly model temporal and interactive features by capturing potential changes in historical interaction patterns through a simple decomposition. We conduct extensive experiments on several dynamic graph datasets to verify the effectiveness and efficiency of TIDFormer. The experimental results demonstrate that TIDFormer excels, outperforming state-of-the-art models across most datasets and experimental settings. Furthermore, TIDFormer exhibits significant efficiency advantages compared to previous Transformer-based methods.

Haipeng Ding, Zhewei Wei, Yuhang Ye

Graph Neural Networks (GNNs) play a pivotal role in graph-based tasks for their proficiency in representation learning. Among the various GNN methods, spectral GNNs employing polynomial filters have shown promising performance on tasks involving both homophilous and heterophilous graph structures. However, The scalability of spectral GNNs on large graphs is limited because they learn the polynomial coefficients through multiple forward propagation executions during forward propagation. Existing works have attempted to scale up spectral GNNs by eliminating the linear layers on the input node features, a change that can disrupt end-to-end training, potentially impact performance, and become impractical with high-dimensional input features. To address the above challenges, we propose "Spectral Graph Neural Networks with Laplacian Sparsification (SGNN-LS)", a novel graph spectral sparsification method to approximate the propagation patterns of spectral GNNs. We prove that our proposed method generates Laplacian sparsifiers that can approximate both fixed and learnable polynomial filters with theoretical guarantees. Our method allows the application of linear layers on the input node features, enabling end-to-end training as well as the handling of raw text features. We conduct an extensive experimental analysis on datasets spanning various graph scales and properties to demonstrate the superior efficiency and effectiveness of our method. The results show that our method yields superior results in comparison with the corresponding approximated base models, especially on dataset Ogbn-papers100M(111M nodes, 1.6B edges) and MAG-scholar-C (2.8M features).

Yanping Zheng, Zhewei Wei, Frank de Hoog et al.

Graph Neural Networks (GNNs) have demonstrated remarkable effectiveness in recommendation systems. However, conventional graph-based recommenders, such as LightGCN, require maintaining embeddings of size $d$ for each node, resulting in a parameter complexity of $\mathcal{O}(n \times d)$, where $n$ represents the total number of users and items. This scaling pattern poses significant challenges for deployment on large-scale graphs encountered in real-world applications. To address this scalability limitation, we propose \textbf{Lighter-X}, an efficient and modular framework that can be seamlessly integrated with existing GNN-based recommender architectures. Our approach substantially reduces both parameter size and computational complexity while preserving the theoretical guarantees and empirical performance of the base models, thereby enabling practical deployment at scale. Specifically, we analyze the original structure and inherent redundancy in their parameters, identifying opportunities for optimization. Based on this insight, we propose an efficient compression scheme for the sparse adjacency structure and high-dimensional embedding matrices, achieving a parameter complexity of $\mathcal{O}(h \times d)$, where $h \ll n$. Furthermore, the model is optimized through a decoupled framework, reducing computational complexity during the training process and enhancing scalability. Extensive experiments demonstrate that Lighter-X achieves comparable performance to baseline models with significantly fewer parameters. In particular, on large-scale interaction graphs with millions of edges, we are able to attain even better results with only 1\% of the parameter over LightGCN.

Libin Qiu, Yuhang Ye, Zhirong Gao et al.

While powerful, the inherent non-determinism of large language model (LLM) agents limits their application in structured operational environments where procedural fidelity and predictable execution are strict requirements. This limitation stems from current architectures that conflate probabilistic, high-level planning with low-level action execution within a single generative process. To address this, we introduce the Source Code Agent framework, a new paradigm built on the "Blueprint First, Model Second" philosophy. Our framework decouples the workflow logic from the generative model. An expert-defined operational procedure is first codified into a source code-based Execution Blueprint, which is then executed by a deterministic engine. The LLM is strategically invoked as a specialized tool to handle bounded, complex sub-tasks within the workflow, but never to decide the workflow's path. We conduct a comprehensive evaluation on the challenging tau-bench benchmark, designed for complex user-tool-rule scenarios. Our results demonstrate that the Source Code Agent establishes a new state-of-the-art, outperforming the strongest baseline by 10.1 percentage points on the average Pass^1 score while dramatically improving execution efficiency. Our work enables the verifiable and reliable deployment of autonomous agents in applications governed by strict procedural logic.

Lu Yi, Runlin Lei, Fengran Mo et al.

Future link prediction on temporal graphs is a fundamental task with wide applicability in real-world dynamic systems. These scenarios often involve both recurring (seen) and novel (unseen) interactions, requiring models to generalize effectively across both types of edges. However, existing methods typically rely on complex memory and aggregation modules, yet struggle to handle unseen edges. In this paper, we revisit the architecture of existing temporal graph models and identify two essential but overlooked modeling requirements for future link prediction: representing nodes with unique identifiers and performing target-aware matching between source and destination nodes. To this end, we propose Cross-Attention based Future Link Predictor on Temporal Graphs (CRAFT), a simple yet effective architecture that discards memory and aggregation modules and instead builds on two components: learnable node embeddings and cross-attention between the destination and the source's recent interactions. This design provides strong expressive power and enables target-aware modeling of the compatibility between candidate destinations and the source's interaction patterns. Extensive experiments on diverse datasets demonstrate that CRAFT consistently achieves superior performance with high efficiency, making it well-suited for large-scale real-world applications.