Time series forecasting is crucial for the development of sophisticated web technologies, driving smarter, more responsive, and data-driven web applications. A key to accurate forecasting lies in effectively capturing the intricate dependencies among different variables (channels). While existing channel-dependent methods have shown strong performance by explicitly modeling inter-channel relationships, they face two critical challenges when applied to high-dimensional datasets with thousands of channels. First, the computational complexity of them grows quadratically with the number of channels, leading to significant scalability issues. Second, attention weights reveal that inter-channel dependencies exhibit both local clusters and global structures, yet current methods fail to disentangle these heterogeneous patterns, resulting in mutual interference and degraded forecasting accuracy. To address these challenges, we propose a novel Channel Reordering-Aligned group Fusion Transformer (CRAFT) for high-dimensional time series forecasting. Specifically, we design an energy-based channel reordering mechanism that reorganizes channels into a minimal-energy state, preserving inherent local-global structures. Building on reordered structure, we introduce a group fusion Transformer that explicitly separates local and global dependencies, significantly reducing computational complexity while enhancing representational clarity. Experiments on high-dimensional datasets demonstrate that CRAFT consistently outperforms baselines, achieving higher forecasting accuracy with lower computational overhead.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
The World Wide Web needs reliable predictive capabilities to respond to changes in user behavior and usage patterns. Time series forecasting (TSF) is a key means to achieve this goal. In recent years, the large language models (LLMs) for TSF (LLM4TSF) have achieved good performance. However, there is a significant difference between pretraining corpora and time series data, making it hard to guarantee forecasting quality when directly applying LLMs to TSF; fine-tuning LLMs can mitigate this issue, but often incurs substantial computational overhead. Thus, LLM4TSF faces a dual challenge of prediction performance and compute overhead. To address this, we aim to explore a method for improving the forecasting performance of LLM4TSF while freezing all LLM parameters to reduce computational overhead. Inspired by in-context learning (ICL), we propose LVICL. LVICL uses our vector-injected ICL to inject example information into a frozen LLM, eliciting its in-context learning ability and thereby enhancing its performance on the example-related task (i.e., TSF). Specifically, we first use the LLM together with a learnable context vector adapter to extract a context vector from multiple examples adaptively. This vector contains compressed, example-related information. Subsequently, during the forward pass, we inject this vector into every layer of the LLM to improve forecasting performance. Compared with conventional ICL that adds examples into the prompt, our vector-injected ICL does not increase prompt length; moreover, adaptively deriving a context vector from examples suppresses components harmful to forecasting, thereby improving model performance. Extensive experiments demonstrate the effectiveness of our approach.
The World Wide Web thrives on intelligent services that rely on accurate time series classification, which has recently witnessed significant progress driven by advances in deep learning. However, existing studies face challenges in domain incremental learning. In this paper, we propose a lightweight and robust dual-causal disentanglement framework (DualCD) to enhance the robustness of models under domain incremental scenarios, which can be seamlessly integrated into time series classification models. Specifically, DualCD first introduces a temporal feature disentanglement module to capture class-causal features and spurious features. The causal features can offer sufficient predictive power to support the classifier in domain incremental learning settings. To accurately capture these causal features, we further design a dual-causal intervention mechanism to eliminate the influence of both intra-class and inter-class confounding features. This mechanism constructs variant samples by combining the current class's causal features with intra-class spurious features and with causal features from other classes. The causal intervention loss encourages the model to accurately predict the labels of these variant samples based solely on the causal features. Extensive experiments on multiple datasets and models demonstrate that DualCD effectively improves performance in domain incremental scenarios. We summarize our rich experiments into a comprehensive benchmark to facilitate research in domain incremental time series classification.
Table reasoning with large language models (LLMs) plays a critical role in building intelligent systems capable of understanding and analyzing tabular data. Despite recent progress, existing methods still face key limitations: their reasoning processes lacks depth and explicit multi-step reasoning, often relying solely on implicit language model understanding. In addition, their reasoning processes suffer from instability, primarily caused by model uncertainty. In this work, we propose STaR, a novel slow-thinking model that can achieve effective and stable table reasoning. To enable effective multi-step reasoning, we design a two-stage training framework consisting of supervised fine-tuning (SFT) warm-up followed by reinforced fine-tuning (RFT). Specifically, in the SFT stage, we construct a high-quality dataset through automatic self-verification. In the RFT stage, we introduce a difficulty-aware reinforcement learning mechanism to further enhance reasoning capabilities. Furthermore, to improve reasoning stability, we introduce trajectory-level uncertainty quantification, which fuses token-level confidence with answer-level consistency, enabling the selection of better reasoning trajectories. Extensive experiments demonstrate that STaR-8B achieves state-of-the-art performance on in-domain benchmarks and exhibits strong generalization to out-of-domain datasets, highlighting its potential for enhancing both effectiveness and stability in table reasoning.
The proliferation of misinformation on video-sharing platforms demands robust detection of video fake news. Existing methods struggle to integrate external world knowledge with internal multimodal cues, limiting their generalization and robustness. In this work, we propose TrueLens, a new framework for video fake news detection that gathers and consolidates dual-level evidence \zznotethrough three primary components, \ie, External Precedent Retriever, Adversarial Contrastor, and Internal Evidential Logic Fusion. At the external level, the External Precedent Retriever first decomposes the query video into textual, visual, and audio queries while leveraging multimodal large language models (MLLMs) to enhance the overall semantic representation. It then applies an entropy-guided multimodal retrieval mechanism to identify the two most similar reference videos from a gallery of real and fake samples, \ie, one real and one fake video. The Adversarial Contrastor integrates these references with the input video through contrastive attention, enhancing contextual reasoning. At the internal level, our Evidential Logic Fusion module aggregates multimodal signals from the Adversarial Contrastor to produce consistent, robust predictions. Extensive experiments on three benchmarks show that the proposed TrueLens consistently surpasses competitive baselines under both temporal and event settings by a clear margin, yielding up to a +21.60% F1 improvement under the event setting and achieving 93.73%, 90.64%, and 98.83% accuracy on the FakeSV, FakeTT, and FVC datasets under the temporal setting. The code for our project is available at https://github.com/JunyiChen-ai/TrueLens.
With the increasing relevance of web mining and content analysis in uncovering mobility patterns from large-scale online data, traffic flow prediction plays a crucial role in proactive urban planning and enhancing the responsiveness of intelligent transportation systems. However, existing traffic flow prediction methods often fail to explicitly capture correlations in continuous multivariate sequences that are naturally suited for trend and periodic pattern extraction by vision models and rely on fixed spatial graphs that neglect the cognitive advantages of granular-ball structures, which limits their ability to model interactions and strengthen spatiotemporal dependencies. To address these challenges, this paper proposes a Cross-domain Mamba framework that integrates Multi-scale Imaging and Granular-ball Computing for traffic flow prediction (MIGC-CMamba). First, a multi-scale sequence imaging method is presented, which converts the original time series into image modality and leverages MambaVision to capture both local and global dependencies. Second, a multi-granularity spatial graph is constructed via granular-ball clustering, which balances global trend representation and local detail preservation. Third, a cross-domain enhancement mechanism adaptively integrates temporal and spatial domains, strengthening spatiotemporal dependencies. Lastly, extensive experiments demonstrate superior performance over state-of-the-art baselines, highlighting how vision-based imaging, cognition-inspired granular-ball modeling, and content-aware mining jointly advance the modeling of spatiotemporal dependencies in traffic flow prediction.
Multivariate time series forecasting is fundamental to web-scale systems. However, frequency-domain forecasters face two structural challenges: (i) frequency bias, where low-amplitude yet informative temporal cues are often submerged in noise and neglected during training; (ii) spectral degradation, where standard neural transformations distort high-amplitude periodic structures, thereby weakening the predictive signal. To address both issues, we propose GFMixer, a decoupled dual-path architecture. GFMixer first applies a Temporal Gradient Block (TGB) to capture low-amplitude information through adaptive selection based on temporal-geometric evidence. Subsequently, a Fourier-Aware Attention Block (FAB) represents high-amplitude, multi-frequency information to mitigate spectral degradation. Finally, both information streams are integrated via time-aligned residual connections within a Gradient Aggregation Block (GAB) for the final forecasting task. Extensive evaluations across seven standard benchmarks (ETT, Weather, Electricity, and Traffic) demonstrate GFMixer's superiority, securing 37 first-place rankings and the lowest average MSE overall. Beyond standalone performance, GFMixer serves as a versatile plug-and-play module that consistently enhances mainstream backbones. Code is available at: https://github.com/superlin30/GFMixer.
In the modern internet, with the proliferation of real-time applications such as online gaming and video conferencing, the timely detection of network jitter has become a critical task in network measurement. Network jitter is defined as the abrupt fluctuations in packet inter-arrival times within network flows, which severely degrade the Quality of Service for these applications. Traditional jitter detection methods primarily focus on macro-level end-to-end or hop-by-hop latency variations, neglecting the fine-grained jitter that occurs within specific flows. In this paper, we present JitterSketch, the first sketch-based algorithm specifically designed for detecting jittery flows. JitterSketch employs a novel three-stage structure to efficiently filter out infrequent and stable flows, thereby identifying and reporting the jittery flows that have the most significant impact on network quality. Extensive experiments demonstrate that JitterSketch achieves an improvement of up to 50 percentage points in both recall and precision rates compared to baseline solutions, while maintaining high processing throughput. Furthermore, we deployed JitterSketch in a QoS simulation system, where it yielded significant improvements in QoS.
Multimodal dialogue emotion recognition seeks to integrate information from text, audio, and video to accurately determine the emotional state of each utterance. However, prevailing approaches often depend on fixed fusion strategies, failing to account for quality variations across modalities. Consequently, noise from less informative modalities can degrade overall performance. Moreover, most models treat utterances as independent units for static classification, overlooking the dynamic evolution of emotions throughout a dialogue. To overcome these limitations, we propose MIDE, a novel approach that leverages mutual information enhancement and dynamic modality slection for multimodal emotion recognition. Our approach dynamically selects high-quality modality pairs for fusion, minimizing interference from noisy or redundant sources. Specifically, it employs mutual information maximization to achieve cross-modal semantic alignment and incorporates an autonomous modality selection mechanism to assess inter-modal compatibility. Furthermore, an emotion transition prediction module, implemented with a Gated Recurrent Unit (GRU), captures temporal emotional dependencies, enabling joint optimization of static emotion classification and dynamic emotion trajectory prediction. Extensive experiments on the IEMOCAP and MELD datasets demonstrate that MIDE significantly surpasses existing models in accuracy and robustness, highlighting its strength in adaptive fusion for complex dialogue scenarios.
Multimodal information extraction (MIE) constitutes a set of essential tasks aimed at extracting structural information from Web texts with integrating images, to facilitate the structural construction of Web-based semantic knowledge. To address the expanding category set including newly emerging entity types or relations on websites, prior research proposed the zero-shot MIE (ZS-MIE) task which aims to extract unseen structural knowledge with textual and visual modalities. However, the ZS-MIE models are limited to recognizing the samples that fall within the unseen category set, and they struggle to deal with real-world scenarios that encompass both seen and unseen categories. The shortcomings of existing methods can be ascribed to two main aspects. On one hand, these methods construct representations of samples and categories within Euclidean space, failing to capture the hierarchical semantic relationships between the two modalities within a sample and their corresponding category prototypes. On the other hand, there is a notable gap in the distribution of semantic similarity between seen and unseen category sets, which impacts the generative capability of the ZS-MIE models. To overcome the above disadvantages, we delve into the generalized zero-shot MIE (GZS-MIE) task and propose the hyperbolic multimodal generative representation learning framework (HMGRL). The variational information bottleneck and autoencoder networks are reconstructed with hyperbolic space for modeling the multi-level hierarchical semantic correlations among samples and prototypes. Furthermore, the proposed model is trained with the unseen samples generated by the decoder, and we introduce the semantic similarity distribution alignment loss to enhance the model's generalization performance. Experimental evaluations on two benchmark datasets underscore the superiority of HMGRL compared to existing baseline methods.
Using Web Mining and Content Analysis to find and understand clues from the massive amount of unstructured text online is very important for predicting future events and providing early risk warnings in important fields like finance and public safety. While Large Language Models (LLMs) exhibit potential in processing and understanding text, current text-based event prediction faces two primary challenges: first, an insufficient utilization of potential information within the text, such as causal relationships and latent associations, and second, limited predictive reliability constrained by issues like the LLM's own ability and hallucinations. To address these challenges, we propose a novel event prediction framework, Bidirectional Reasoning with Self-Correction (BRSC). BRSC comprises two complementary reasoning dimensions: temporal deductive reasoning, which analyzes the trajectory of historical events along the timeline to enable accurate trend extrapolation, and synchronic associative reasoning, which deeply mines details and latent connections from documents within a specific time window to extended semantic information. In addition, we use a self-correction mechanism that identifies and rectifies potential hallucinations and errors during the reasoning process. Extensive experiments on international relations event prediction demonstrate that BRSC achieves significant improvements over several leading LLM-based methods.
Identifying central claims from long documents is a fundamental yet challenging task in automated fact-checking pipelines. Manual extraction at the document level is costly and requires domain expertise, while existing automatic methods for claim extraction and evaluation tend to overlook critical dimensions that human fact-checkers consider essential for determining claim quality. In this paper, we propose a novel Human Value-Aligned framework that enables zero-shot document-level claim extraction and evaluation by aligning with expert preferences. We first elicit a structured set of Human Value Alignment (HVA) dimensions from expert annotations and incorporate them into prompt design, instructing large language models (LLMs) to extract high-quality claims that align with expert values. To assess the quality of extracted claims, we further introduce an LLM-based automatic evaluator that scores claims across HVA dimensions and quantifies alignment with expert-written claims. Furthermore, we propose a multi-level agreement metric to evaluate the reliability of automatic HVA evaluator. Experiment results show that our method significantly improves central claim extraction performance, achieving state-of-the-art chrF and P@N scores. Moreover, the proposed HVA evaluator achieves high agreement with human judgments and offers interpretable dimension-level assessments of extracted claims. The HVA framework establishes a reliable and scalable way for human-aligned document-level claim extraction and evaluation in real-world scenarios.
Predictive modeling on web-scale tabular data presents significant scalability challenges for industrial applications, often involving billions of instances and hundreds of heterogeneous numerical features. The inherent complexities of these features—characterized by anisotropy, heavy-tailed distributions, and non-stationarity—not only impose bottlenecks on the training efficiency and scalability of mainstream models like Gradient Boosting Decision Trees (GBDTs), but also compel practitioners into laborious, inefficient, and expert-dependent manual feature engineering. To systematically address this challenge, we introduce KMLP, a novel hybrid deep architecture. KMLP synergistically integrates a shallow Kolmogorov-Arnold Network (KAN) as a front-end with a Gated Multilayer Perceptron (gMLP) as the backbone. The KAN front-end leverages its learnable activation functions to automatically model complex non-linear transformations for each input feature in an end-to-end manner, thereby automating feature representation learning. Subsequently, the gMLP backbone efficiently captures high-order interactions among these refined representations. Extensive experiments on multiple public benchmarks and an ultra-large-scale industrial web dataset with billions of samples demonstrate that KMLP achieves state-of-the-art (SOTA) performance. Crucially, our findings reveal that KMLP's performance advantage over strong baselines like GBDTs becomes more pronounced as the data scale increases. This validates KMLP as a scalable and adaptive deep learning paradigm, offering a promising path forward for modeling large-scale, dynamic web tabular data.
Knowledge tracing (KT) aims to personalize online education on large-scale web-based platforms by modeling students' evolving knowledge states from their interaction sequences. However, most KT models rely on a single encoder architecture (e.g., self-attention or RNN), with fixed inductive biases that fails to capture the diversity of learning behaviors. Specifically, student learning unfolds across multiple timescales, and interaction sequences contain diverse frequency components ranging from short-term variations to long-term trends. Our data-driven analysis reveals that existing encoders exhibit characteristic frequency biases (e.g., self-attention tends to emphasize low-frequency patterns), highlighting the limitations of any single architecture. To address this problem, we propose FA-KT, a frequency-aware mixture of heterogeneous experts framework. FA-KT combines self-attention, Mamba, CNN, and LSTM experts, each with complementary frequency biases. A frequency-aware router analyzes each sequence's frequency characteristics and adaptively combines experts to create dynamic, personalized encoders for individual students. Across five benchmark datasets, FA-KT consistently outperforms 20 strong KT baselines in predicting future performance. Code is available at https://pykt.org/.
Interlingual subtitling, which translates subtitles of visual media into a target language, is essential for entertainment localization but has not yet been explored in machine translation. Although Large Language Models (LLMs) have significantly advanced the general capabilities of machine translation, the distinctive characteristics of subtitle texts pose persistent challenges in interlingual subtitling, particularly regarding semantic coherence, pronoun and terminology translation, and translation expressiveness. To address these issues, we present Hermes, an LLM-based automated subtitling framework. Hermes integrates three modules: Speaker Diarization, Terminology Identification, and Expressiveness Enhancement, which effectively tackle the above challenges. Experiments demonstrate that Hermes achieves state-of-the-art diarization performance and generates expressive, contextually coherent translations, thereby advancing research in interlingual subtitling.
Generalized Category Discovery (GCD) is a critical task in open-world computing scenarios, aiming to automatically classify partially labeled data by recognizing both known and novel categories. However, existing GCD methods usually suffer from inherent bias toward known categories due to the exclusive pre-training on them and the absence of labeled data of novel categories. This bias can lead to significant misclassification and clustering errors for novel categories. Although recent approaches leverage pseudo-label training and contrastive learning to address this, they still lack explicit supervision to disentangle novel and known categories, resulting in performance bottlenecks. To address these limitations, we propose an Expectation-Maximization-driven Contrastive Disentanglement (EMCD) framework designed to explicitly disentangle novel and known categories. We particularly formulate the identification of novel categories as a latent variable estimation problem. Specifically, it incorporates an EM-disentangling regularization to softly identify novel category samples and a consistency regularization to enhance generalization. In addition, we leverage dual contrastive constraints, including a cluster-sample contrastive constraint and a sample-sample contrastive contrastive, to pull close samples of novel categories while pushing apart ambiguous samples near decision boundaries. Empirical results on 3 commonly used datasets demonstrate that our model is effective and outperforms previous state-of-the-art methods. Our code is available at https://github.com/YWY-only/EMCD.
The rapid spread of online misinformation poses a serious threat to public trust and social stability, making automatic fake news detection increasingly critical. In this work, we propose a novel retrieval-augmented fake news detection framework, Triple-R (Rewriting, Retrieval, and iterative Refinement), which emphasizes optimizing the retrieval query. Unlike prior retrieval-augmented methods that adapt either the retriever or the verification module, often overlooking the gap between news text and the evidence needed for verification, our approach focuses on adapting the search query to retrieve the most relevant evidence. Specifically, we employ a small language model as a trainable query rewriter, optimized via reinforcement learning with feedback from a frozen LLM-based fake news detector, to transform the original news text into effective retrieval queries. To further enhance evidence relevance, we introduce an iterative query refinement mechanism, which progressively updates rewritten queries based on previously retrieved results. Finally, the original news text and the evidence retrieved through refined queries are integrated for verification. Experiments on two real-world datasets demonstrate consistent improvements, validating the effectiveness of our approach.
Zero-shot stance detection (ZSSD) aims to classify stances towards previously unseen targets without direct supervision on those topics during training. While recent approaches have explored various strategies, they often suffer from limited linguistic diversity or unstable semantic representations, restricting generalization to new domains. To address these challenges, we propose DPSD , a dynamic framework that integrates LLM-assisted data augmentation, multi-granularity feature fusion with contrastive learning, and adaptive prototype updating. Our method enriches the training data with both diverse targets and stylistically varied texts. A gate-controlled fusion mechanism combines deep contextualized features from BERT with shallow lexical patterns via TF-IDF, while contrastive learning refines the feature space by pulling similar instances closer and pushing dissimilar ones apart, thereby improving representation discriminability. Furthermore, we introduce a sliding-window prototype pool that dynamically maintains class-specific prototypes while preserving historical knowledge, ensuring stable and interpretable inference over time. We also incorporate LLM-calibrated semantic similarity as an auxiliary scorer for controlled reasoning. Experimental results on three benchmark datasets -SEM16, P-Stance, and VAST - show that DPSD achieves strong performance in various zero-shot settings, especially in cross-dataset and unseen target scenarios.
The rapid advancement of multimodal large language models has enabled agents to operate mobile devices by directly interacting with graphical user interfaces, opening new possibilities for mobile automation. However, real-world mobile tasks are often complex and allow for multiple valid solutions. This contradicts current mobile agent evaluation standards: offline static benchmarks can only validate a single predefined ''golden path'', while online dynamic testing is constrained by the complexity and non-reproducibility of real devices, making both approaches inadequate for comprehensively assessing agent capabilities. To bridge the gap between offline and online evaluation and enhance testing stability, this paper introduces a novel graph-structured benchmarking framework. By modeling the finite states observed during real-device interactions, it achieves static simulation of dynamic behaviors. Building on this, we develop ColorBench, a benchmark focused on complex long-horizon tasks. It supports evaluation of multiple valid solutions, subtask completion rate statistics, and atomic-level capability analysis. ColorBench contains 175 tasks (74 single-app, 101 cross-app) with an average length of over 13 steps. Each task includes at least two correct paths and several typical error paths, enabling quasi-dynamic interaction.
Temporal Point Processes (TPPs) have been widely used for modeling event sequences on the Web, such as user reviews, social media posts, and online transactions. However, traditional TPP models often struggle to effectively incorporate the rich textual descriptions that accompany these events, while Large Language Models (LLMs), despite their remarkable text processing capabilities, lack mechanisms for handling the temporal dynamics inherent in Web-based event sequences. To bridge this gap, we introduce Language-TPP, a unified framework that seamlessly integrates TPPs with LLMs for enhanced Web event sequence modeling. Our key innovation is a novel temporal encoding mechanism that converts continuous time intervals into specialized byte-tokens, enabling direct integration with standard language model architectures for TPP modeling without requiring TPP-specific modifications. This approach allows Language-TPP to achieve state-of-the-art performance across multiple TPP benchmarks, including event time prediction and type prediction, on real-world Web datasets spanning e-commerce reviews, social media and online Q&A platforms. More importantly, we demonstrate that our unified framework unlocks new capabilities for TPP research: incorporating temporal information improves the quality of generated event descriptions, as evidenced by enhanced ROUGE-L scores, and better aligned sentiment distributions. Through comprehensive experiments, including qualitative analysis of learned distributions and scalability evaluations on long sequences, we show that Language-TPP effectively captures both temporal dynamics and textual patterns in Web user behavior, with important implications for content generation, user behavior understanding, and Web platform applications. Code is available at https://github.com/qykong/Language-TPP.