Search is not available for this dataset
paper_name
string
paper_url
string
paper_authors
list
paper_abstract
string
paper_code
string
conf
string
Causal Structure-guided Distributionally Robust Optimization under Domain Shifts
https://doi.org/10.1145/3770855.3818154
[ "Seonggyeom Kim", "Eunjung Choi", "Dong-Kyu Chae" ]
Domain generalization (DG) aims to learn predictive models from multiple source domains that maintain performance on unseen target domains. Distributionally robust optimization (DRO) addresses distribution shift by minimizing the worst-case risk over an uncertainty set of plausible test distributions. However, if this ...
https://github.com/gyeomo/CS-DRO
KDD2026
Spectral Integrated Gradients for Coarse-to-Fine Feature Attribution
https://doi.org/10.1145/3770855.3817830
[ "Soyeon Kim", "Seongwoo Lim", "Kyowoon Lee", "Jaesik Choi" ]
Integrated Gradients (IG) is a widely adopted feature attribution method that satisfies desirable axiomatic properties. However, the choice of integration path significantly affects the quality of attributions, and the standard straight-line path introduces all input features simultaneously, often accumulating noisy gr...
https://github.com/leekwoon/sig/
KDD2026
SAGEO Arena: A Realistic Environment for Evaluating Search-Augmented Generative Engine Optimization
https://doi.org/10.1145/3770855.3818146
[ "Sunghwan Kim", "Wooseok Jeong", "Serin Kim", "Sangam Lee", "Dongha Lee" ]
Search-Augmented Generative Engines (SAGE) have emerged as a new paradigm for information access, bridging web-scale retrieval with generative capabilities to deliver synthesized answers. This shift has fundamentally reshaped how web content gains exposure online, giving rise to Search-Augmented Generative Engine Optim...
#
KDD2026
Drift-Aware Memory-Augmented Spatio-Temporal Graph Attention for Industrial Anomaly Detection
https://doi.org/10.1145/3770855.3818125
[ "Yooshin Kim", "Donghoon Shin" ]
Industrial Control Systems (ICSs) increasingly face cyber-physical attacks, making reliable anomaly detection essential. However, ICS data are high-dimensional multivariate time series with complex spatio-temporal dependencies and non-stationary device behaviors. Existing methods typically encode spatial relations with...
#
KDD2026
inversedMixup: Data Augmentation via Inverting Mixed Embeddings
https://doi.org/10.1145/3770855.3817781
[ "Fanshuang Kong", "Richong Zhang", "Qiyu Sun", "Zhijie Nie", "Ting Deng", "Chunming Hu" ]
Mixup generates augmented samples by linearly interpolating inputs and labels with a controllable ratio. However, since it operates at the latent embedding level, the resulting samples are not human-interpretable. In contrast, LLM-based augmentation methods produce sentences via prompts at the token level, yielding rea...
https://github.com/pypi1412/inversedMixup
KDD2026
Learned Lossless Image Compression with Interleaved Parallel Inference and Irregular Causal Reasoning
https://doi.org/10.1145/3770855.3817757
[ "Lingdu Kong", "Xiaochun Yang", "Shuo Li", "Tieying Li", "Bin Wang", "Chunhui Shen", "Xiang Wang", "Feibo Li" ]
Lossless image compression typically leverages already decoded pixels to reconstruct the undecoded ones. Existing learned lossless image compression methods typically adopt either pixel-by-pixel or block-based decoding strategies. Pixel-by-pixel decoding fully exploits spatial priors but introduces high computational o...
#
KDD2026
AlphaOPT: Formulating Optimization Programs with Self-Improving LLM Experience Library
https://doi.org/10.1145/3770855.3817974
[ "Minwei Kong", "Ao Qu", "Xiaotong Guo", "Wenbin Ouyang", "Chonghe Jiang", "Han Zheng", "Yining Ma", "Dingyi Zhuang", "Yuhan Tang", "Junyi Li", "Shenhao Wang", "Haris N. Koutsopoulos", "Hai Wang", "Cathy Wu", "Jinhua Zhao" ]
Optimization modeling underlies critical decision-making across industries, yet remains difficult to automate: natural-language problem descriptions must be translated into precise mathematical formulations and executable solver code. Existing LLM-based approaches typically rely on brittle prompting or costly retrainin...
https://github.com/Minw913/AlphaOPT
KDD2026
A More Accurate Algorithm Comparison through A/B Testing using Offline Evaluation Methods
https://doi.org/10.1145/3770855.3817659
[ "Koki Konishi", "Masataka Ushiku", "Yuta Saito" ]
A/B testing is the gold standard for selecting better algorithms in online services. While offline evaluation has attracted attention as a safer alternative due to the high experimental costs and the potential risk of degrading user experience and revenue in A/B testing, it is widely recognized that the estimation accu...
#
KDD2026
Learning to Doubt: Forgetting Aware Learning for Neural Networks
https://doi.org/10.1145/3770855.3818180
[ "Awanish Kumar", "Soumyadeep Ghosh", "Akshita Sharma", "Rahul Gupta" ]
Modern neural networks are often miscalibrated, assigning high confidence to predictions they have not learned stably, which leads to overconfident errors under noise, imbalance, and distribution shift. We propose Forgetting-Aware Learning (FAL) using a simple ranking regularizer that leverages forgetting events during...
#
KDD2026
Semi-Supervised Text-Attributed Graph Distillation
https://doi.org/10.1145/3770855.3817685
[ "Yurui Lai", "Samir Moustafa", "Renchi Yang", "Tsz Nam Chan" ]
Text-Attributed Graphs (TAGs) have emerged as an expressive data model for integrating graph topology with rich textual semantics. Existing representation learning methods over TAGs suffer from severe scalability bottlenecks, particularly together with Large Language Models (LLMs). While data distillation offers a prom...
#
KDD2026
Efficient Test-Time Scaling for LLM-based Time Series Forecasting
https://doi.org/10.1145/3770855.3818201
[ "Xuan-May Le", "Minh-Tuan Tran", "Ling Luo", "Uwe Aickelin", "Dinh Q. Phung", "Trung Le" ]
Long-term time series forecasting benefits from preserving global structure such as trends and seasonality. Recent LLM-based forecasters often improve accuracy through test-time scaling (e.g., iterative refinement), but these methods are computationally expensive and increasingly prone to global-shape mismatch as the p...
https://github.com/xuanmay2701/SCALER
KDD2026
Entrywise Error Bounds for Spectral Ranking with Semi-Random Adversaries
https://doi.org/10.1145/3770855.3818135
[ "Dongmin Lee", "Anuran Makur", "Japneet Singh" ]
Bradley-Terry-Luce (BTL) model estimation is a well-established strategy to rank a collection of items given a dataset of pairwise comparisons. Although the theoretical performance of BTL estimation methods, such as spectral and maximum likelihood estimation, is well studied in the regime of uniformly sampled graphs, g...
#
KDD2026
Node4All: Learning Node Representation Beyond Datasets
https://doi.org/10.1145/3770855.3817828
[ "Dooho Lee", "Jaemin Yoo" ]
Node representation learning has advanced rapidly, yet most existing methods rely on per-dataset training and hyperparameter tuning. This dataset-specific optimization comes from the difficulty of designing reusable graph models that generalize across diverse graph datasets. In this work, we introduce Node4All, a node ...
https://github.com/dooho00/node4all
KDD2026
RiskBound: Risk-Aware Boundary-Guided Portfolio Optimization via Action Space Reshaping
https://doi.org/10.1145/3770855.3817928
[ "Hosung Lee", "U. Kang" ]
How can we build investment portfolios that remain robust under shifting market regimes and time-varying risk exposure? Portfolio optimization is a crucial problem in the financial domain, involving periodic asset allocation decisions to maximize long-term returns under risk. Recent advances have highlighted the effect...
#
KDD2026
Learn When and Where to Connect: Adaptive Virtual Nodes for Dynamic Message Passing on Graphs
https://doi.org/10.1145/3770855.3818013
[ "Jaejun Lee", "Joyce Jiyoung Whang" ]
While Virtual Nodes (VNs) are often utilized in Message Passing Neural Networks (MPNNs) to facilitate effective message passing, existing VN-based methods have limitations, such as constraining all nodes to connect to the same number of VNs, fixing the connections before applying MPNNs, and connecting a node to a VN in...
#
KDD2026
TiRano: Tensorized Relation-aware Temporal Reasoning for Accurate Knowledge Graph Completion
https://doi.org/10.1145/3770855.3817784
[ "SeungJoo Lee", "Yong-chan Park", "U. Kang" ]
Given a partially observed Temporal Knowledge Graph (TKG), how can we accurately predict missing entities? Unlike static knowledge graphs, TKGs encode facts within temporal contexts, requiring models to reason over both graph structure and time. However, existing TKGC approaches often sample neighbors solely based on t...
#
KDD2026
Can Structural Cues Save LLMs? Evaluating Language Models in Massive Document Streams
https://doi.org/10.1145/3770855.3817994
[ "Yukyung Lee", "Yebin Lim", "Woojun Jung", "Wonjun Choi", "Susik Yoon" ]
Evaluating language models in streaming environments is critical, yet underexplored. Existing benchmarks either focus on single complex events or provide curated inputs for each query, and do not evaluate models under the conflicts that arise when multiple concurrent events are mixed within the same document stream. We...
#
KDD2026
NaviRAG: Learning to Navigate Knowledge Graphs for Retrieval-Augmented Generation
https://doi.org/10.1145/3770855.3817935
[ "Jinghong Lei", "Wang Kun", "Zhigang Chen", "Zhao Xin" ]
Retrieval-Augmented Generation (RAG) has become a fundamental paradigm for enhancing Large Language Models (LLMs) with external knowledge. However, while recent structure-augmented approaches organize documents into graphs to improve information access, their retrieval strategies remain largely static, relying on simil...
https://github.com/CkingEW/NaviRAG
KDD2026
VaaWIT: Visual-Aware Adaptation of Large Language Models for Multilingual Web Image Translation
https://doi.org/10.1145/3770855.3817631
[ "Bo Li", "Ronghao Chen", "Ningyuan Deng", "Huacan Wang", "Shaolin Zhu", "Lijie Wen" ]
Translating text embedded in Web images is crucial for improving content accessibility and cross-lingual information retrieval, particularly within social media and e-commerce domains. Although Large Vision-Language Models (LVLMs) have advanced multimodal understanding, applying them to Web image translation remains ch...
#
KDD2026
Beyond One-Size-Fits-All: Adaptive Subgraph Denoising for Zero-Shot Graph Learning with Large Language Models
https://doi.org/10.1145/3770855.3818174
[ "Fengzhi Li", "Liang Zhang", "Yuan Zuo", "Ruiqing Zhao", "YanSong Liu", "Yunfei Ma", "Fanyu Meng", "Junlan Feng" ]
Graph-based tasks in the zero-shot setting remain a significant challenge due to data scarcity and the inability of traditional Graph Neural Networks (GNNs) to generalize to unseen domains or label spaces. While recent advancements have transitioned toward leveraging Large Language Models (LLMs) as predictors to enhanc...
https://github.com/mysteriouslfz/GraphSSR
KDD2026
LatentCRS: A Variational EM Framework for Bridging Semantics and Behavior in LLM-based Conversational Recommendation
https://doi.org/10.1145/3770855.3818202
[ "Guanrong Li", "Kuo Tian", "Jinnan Qi", "Qinghan Fu", "Zhen Wu", "Rui Xia", "Xinyu Dai" ]
Conversational Recommender Systems (CRS) powered by Large Language Models (LLMs) enable users to articulate explicit and dynamic preferences, overcoming the limitations of fixed templates. However, despite their superior semantic proficiency, LLMs have not yet achieved corresponding improvements in recommendation accur...
#
KDD2026
Expand More, Shrink Less: Shaping Effective-Rank Dynamics for Dense Scaling in Recommendation
https://doi.org/10.1145/3770855.3818049
[ "Guoming Li", "Shangyu Zhang", "Junwei Pan", "Wentao Ning", "Jin Chen", "Gengsheng Xue", "Chao Zhou", "Shudong Huang", "Haijie Gu", "Menglin Yang" ]
Scaling recommendation models is a central challenge in recommender systems. Recently, RankMixer has emerged as an effective solution, operating on a unified token representation and alternating between token mixing and per-token feedforward networks (P-FFNs) to achieve scalable performance. However, RankMixer suffers ...
https://github.com/vasile-paskardlgm/RankElastor
KDD2026
LSAR: Sparse Lexical Representation Learning for Efficient and Interpretable Audio Retrieval
https://doi.org/10.1145/3770855.3818190
[ "Haoyue Li", "Yuzhe Bai", "Li Niu" ]
As Multimodal Large Language Models (MLLMs) expand the scope of retrieval-augmented generation, recommendation, and multimedia search, audio retrieval is expected to become a dependable retrieval component. Yet existing systems struggle to reconcile lexical precision, non-verbal acoustic evidence, and efficient, transp...
#
KDD2026
LATTE: Learning Adaptive Segmentation for Efficient and Effective Trajectory Similarity Learning
https://doi.org/10.1145/3770855.3817880
[ "Jialiang Li", "Hua Lu", "Tiantian Liu", "Zhichen Lai", "Pengfei Li" ]
Trajectory similarity learning (TSL) models represent variable-length trajectories as fixed-size vectors, for which the similarity is much easier to compute. Recent TSL models often map-match raw GPS trajectories onto road networks to exploit richer information. Although effective, such models rely on heavy modeling st...
#
KDD2026
Collaborative Knowledge Extraction and Integration for Graph Domain Incremental Learning
https://doi.org/10.1145/3770855.3817613
[ "Jialu Li", "Yu Wang", "Wanyu Lin", "Pengfei Zhu" ]
Graph Domain Incremental Learning (GDIL) aims to acquire knowledge from a continuous stream of graph domains while mitigating catastrophic forgetting. While parameter-isolation methods leveraging graph parameter-efficient adaptation show promise, prompt-based techniques struggle to adapt to GDIL, and low-rank adaptatio...
#
KDD2026
FreqCore: A Frequency Domain Perspective on Coreset Selection
https://doi.org/10.1145/3770855.3817832
[ "Jiazhe Li", "Chenhe Hao", "Weiying Xie", "Jitao Ma", "Daixun Li", "Xin Zhang", "Leyuan Fang" ]
Coreset selection aims to select an informative subset that achieves performance comparable to the whole dataset. However, existing methods are often susceptible to spurious correlations present in high-frequency domain, consequently leading to inaccurate sample importance estimation and suboptimal coresets. To address...
#
KDD2026
Revisiting Graph Autoencoders as Implicit Contrastive Learners
https://doi.org/10.1145/3770855.3817903
[ "Jintang Li", "Ruofan Wu", "Yuchang Zhu", "Huizhe Zhang", "Zulun Zhu", "Liang Chen" ]
Graph autoencoders (GAEs) and graph contrastive learning (GCL) are two major paradigms for self-supervised representation learning on graphs, yet they are often studied in isolation and treated as fundamentally different approaches. In this work, we revisit GAEs through the lens of contrastive learning and show that bo...
#
KDD2026
Bridging Front-Door Adjustment and Information Bottleneck for Identifiable Causal Representations
https://doi.org/10.1145/3770855.3818026
[ "Jue Li", "Yuhua Qian", "Jieting Wang", "Saixiong Liu", "Honghong Cheng" ]
The Information Bottleneck (IB) essentially constructs Z as a learnable intermediate variable for the prediction task by balancing the information between the learned compressed representation Z and the prediction target Y. However, when there is unobserved confounding, the mediators learned by IB are often contaminate...
https://github.com/lijue688/FDA-IB
KDD2026
A Product Manifold Method for Feature Selection
https://doi.org/10.1145/3770855.3817798
[ "Mao Li", "Zhilong Mi", "Yingpeng Du", "Yifan Cao", "Qingcai He", "Ziqiao Yin", "Binghui Guo", "Zhu Sun" ]
Feature selection is indispensable for mitigating overfitting and reducing feature redundancy in high-dimensional scenarios. However, most existing approaches rely on unstable overall separability and sample-wise geometry, thereby neglecting the stable class-specific discriminative structures and leading to poor perfor...
#
KDD2026
Dual-Difficulty Curriculum Learning for Direct Preference Optimization
https://doi.org/10.1145/3770855.3817815
[ "Mengyang Li", "Haozhan Geng", "Zhong Zhang", "Shuang Liu" ]
Curriculum learning enhances Direct Preference Optimization (DPO) for aligning Large Language Models (LLMs), yet existing methods rely on a one-dimensional view of difficulty. In this work, we reframe alignment difficulty as a two-dimensional space spanned by Prompt Complexity (PC) and Pairwise Distinguishability (PD),...
#
KDD2026
DeMIC: Decentralized Meta In-Context Learning with Refiner-Guided Adaptation
https://doi.org/10.1145/3770855.3818159
[ "Mingyi Li", "Xiao Zhang", "Zengzhe Chen", "Jiawei Zhang", "Yuan Yuan", "Wei Guo", "Fuzhen Zhuang", "Dongxiao Yu" ]
In-Context Learning (ICL) empowers Large Language Models to adapt to novel tasks, presenting a promising solution for privacy-preserving distributed scenarios. However, relying exclusively on inference-time ICL suffers from inherent instability. While fine-tuning models on specific tasks could mitigate these issues, as...
#
KDD2026
From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Retrieval-Augmented Generation Agents Development
https://doi.org/10.1145/3770855.3817782
[ "Muzhi Li", "Jinhu Qi", "Yihong Wu", "Minghao Zhao", "Liheng Ma", "Yifan Li", "Xinyu Wang", "Zhenghan Tai", "Zixing Song", "Yingxue Zhang", "Ho-fung Leung", "Irwin King" ]
Retrieval-augmented generation (RAG) agent development is hindered by the lack of executable ground-truth agent-environment interaction trajectories. Existing datasets provide questions, answers, and evidence, but lack fine-grained supervision for retriever invocation, dynamic planning, and stepwise decision-making. Re...
#
KDD2026
LCATS: LLM-Guided Constraint-Aware Tabular Data Synthesis
https://doi.org/10.1145/3770855.3818181
[ "Qing Li", "Yanyan Shen", "Qibin Zheng", "Yi Liu", "Xingchun Diao" ]
Tabular data synthesis must satisfy two complementary quality criteria. At the global level, synthetic records must faithfully reproduce the statistical characteristics of the original dataset, including marginal distributions and inter-variable correlations. At the local level, each individual record must satisfy doma...
https://github.com/SJTU-DMTai/LCATS
KDD2026
ReCATS: Replay-Free Continual Anomaly Detection for Non-Stationary Multivariate Time Series
https://doi.org/10.1145/3770855.3817985
[ "Qiuyang Li", "Qian Ma", "Zhongming Yao", "Shikai Guo", "Yu Gu", "Ge Yu", "Tianyi Li" ]
Multivariate time series anomaly detection is critical in safety-critical domains such as industrial monitoring and financial systems. However, real-world time series are inherently non-stationary, with evolving data distributions driven by changing operational regimes and system dynamics. As a result, most existing me...
https://github.com/Li-Qiuyang/ReCATS
KDD2026
Graph Joint Representation Learning for Sound and Scalable Instantiation
https://doi.org/10.1145/3770855.3818189
[ "Tian Li", "Xueli Liu", "Bowen Dong", "Binrong Liu" ]
Knowledge graph reasoning is a fundamental problem in artificial intelligence, where rule-based methods provide strong interpretability but suffer from exponential symbolic search complexity, while existing neuro-symbolic approaches struggle to model complex Horn rules with rich topologies. We propose Rule–Graph Joint ...
#
KDD2026
Adaptive Prototypical Contrastive Learning for Time Series Clustering
https://doi.org/10.1145/3770855.3817773
[ "Wei Li" ]
Unsupervised analysis of time series is pivotal for IoT, healthcare, and industrial monitoring, yet standard methods face a fundamental dilemma: representation learning relies on instance discrimination, while clustering requires semantic grouping. Worse still, existing deep clustering methods depend on a pre-defined c...
#
KDD2026
Effective Intra-Inter Interaction Learning for Relational Tables
https://doi.org/10.1145/3770855.3817787
[ "Weichen Li", "Ken Zhong", "Zheng Wang", "Li Pan", "Jianhua Li" ]
Relational table learning has recently emerged as an important research direction for modeling multiple tables connected through primary key-foreign key (PK-FK) relationships. Despite recent advances, a principled modeling framework tailored to this task remains underexplored. In this paper, we propose Intra-Inter Rela...
https://github.com/W1nterFloW/InRTL
KDD2026
Lasso: Accurate and Efficient Detection of Long-Lived Sparse Items in High-Speed Data Streams
https://doi.org/10.1145/3770855.3818048
[ "Weihe Li", "Jiawei Huang", "Zhaoyi Li", "Tianyue Chu", "Qichen Su", "Jianxin Wang" ]
In high-speed data streams, identifying long-lived (also referred to as persistent) sparse items is critical, as such patterns may indicate stealthy or low-rate threats yet remain largely underexplored. Although recent studies have begun to examine this problem, existing approaches either suffer from low lookup accurac...
#
KDD2026
BiVCoder: A Multi-Agent Framework for Code Generation via Bidirectional Code-Test Diagnosis
https://doi.org/10.1145/3770855.3818107
[ "Xiaoyang Li", "Jinhao Dong", "Wenhang Shi", "Wei Lu", "Xiaoyong Du" ]
Large Language Models (LLMs) have demonstrated remarkable potential in automated code generation. However, existing test-driven code generation and refinement frameworks are often hindered by the tests' quality: they typically treat self-generated tests as ground truth, leading to ineffective debugging loops where code...
#
KDD2026
Modality-Agnostic Zeroth-Order LoRA Fine-Tuning for Black-Box Prompt Optimization
https://doi.org/10.1145/3770855.3817738
[ "Xingchen Li", "Jia Zhang", "Tianxing Man", "Wenkang Wang", "Bin Gu" ]
Recent progress in large-scale models with generative task capabilities allows users to produce high-quality content directly from text prompts. These contents can be further improved through prompt optimization, which is usually treated as a black-box optimization problem since gradients are inaccessible in the Model-...
https://github.com/xcli23/ZOLO
KDD2026
Causal Abstraction Learning for Multi-Modal Grounded Planning
https://doi.org/10.1145/3770855.3817995
[ "Xinshu Li", "Shiyi Yang", "Ziqi Xu", "Feng Xia", "Quan Z. Sheng", "Lina Yao" ]
Recent advances in multimodal embodied agents have enabled long-horizon planning in visually rich environments via natural language. Yet, their generalization remains brittle when task instructions deviate from familiar examples, exposing a reliance on surface imitation rather than structural understanding. We propose ...
#
KDD2026
Training a Generalist Hallucination Detector across Multiple Domains via Adaptive Layer Aggregation
https://doi.org/10.1145/3770855.3817792
[ "Xinyi Li", "Zhen Fang", "Yadan Luo", "Yongxin Deng", "Sean Du", "Shanshan Ye", "Ling Chen" ]
Hallucination detection is essential for the reliable deployment of large language models (LLMs) in safety-critical applications, yet existing detectors often degrade under real-world domain shifts. To formalize this challenge, we introduce Multi-Domain Generalizable Hallucination Detection (MGHD), where a detector is ...
#
KDD2026
AeroWF: A Geometric Spectral-Temporal Dual-Stream Learning Framework for Aerodrome Weather Forecasting
https://doi.org/10.1145/3770855.3817822
[ "Xinyu Li", "Yunyi Huang", "Quan Fang", "Yang Yang", "Can Zhao", "Kaiquan Cai" ]
Accurate and reliable aerodrome weather forecasting is essential for aerodrome safety and efficiency. However, existing approaches often fail to effectively capture the multi-scale dynamics and structural heterogeneity inherent in airport meteorological data, which involves variable numbers of runways, entangled high-f...
https://github.com/MKC-Lab/AeroWF
KDD2026
Perturbation Effects on Robustness and Individual Fairness
https://doi.org/10.1145/3770855.3817848
[ "Xuran Li", "Hao Xue", "Peng Wu", "Xingjun Ma", "Zhen Zhang", "Huaming Chen", "Flora D. Salim" ]
Deep neural networks are vulnerable to adversarial perturbations that can simultaneously degrade prediction robustness and individual fairness across diverse application settings. However, existing evaluation protocols typically assess these dimensions in isolation, thereby obscuring critical failure modes. To bridge t...
https://github.com/Xuran-LI/RIFair
KDD2026
Mitigating Anomaly Hallucination: A Model-Agnostic Framework for Unsupervised Anomaly Detection on Dynamic Graphs
https://doi.org/10.1145/3770855.3817610
[ "Yingxuan Li", "Yuanyuan Xu", "Xuemin Lin", "Ying Zhang" ]
Temporal graph neural networks (T-GNNs) are powerful for representation learning on continuous-time dynamic graphs (CTDGs), but naively applying them to unsupervised anomaly detection is often unreliable. The key challenge is learning normal dynamics while being exposed to unlabeled anomalies, which destabilizes optimi...
#
KDD2026
AGMark: Attention-Guided Dynamic Watermarking for Large Vision-Language Models
https://doi.org/10.1145/3770855.3817700
[ "Yue Li", "Xin Yi", "Dongsheng Shi", "Yongyi Cui", "Gerard de Melo", "Linlin Wang" ]
Watermarking has emerged as a pivotal solution for content traceability and intellectual property protection in large vision language models (LVLMs). However, vision-agnostic watermarks may introduce visually irrelevant tokens and disrupt visual grounding by enforcing indiscriminate pseudo-random biases. Additionally, ...
#
KDD2026
Mitigating Neuro-Symbolic Reasoning Shortcuts with Data-Driven Knowledge Augmentation
https://doi.org/10.1145/3770855.3817988
[ "Yu-Feng Li", "Xiao-Wen Yang", "Wen-Da Wei", "Jie-Jing Shao", "Lan-Zhe Guo" ]
Recent advancements in neuro-symbolic learning (NeSy) have shown significant promise in integrating deep learning with symbolic reasoning, offering both interpretability and generalization. However, the prevalence of reasoning shortcuts, where the NeSy system predicts incorrect intermediate concepts while maintaining h...
#
KDD2026
KGA-LM: Representation-Level Grounding for Conversational Search over Knowledge Graphs
https://doi.org/10.1145/3770855.3817824
[ "Yunfei Li", "Chengfei Liu", "Rui Zhou", "Zhiyu Xu", "Siyu Zhang", "Jin Wei" ]
Integrating structured knowledge graphs (KGs) with Large Language Models (LLMs) is essential for trustworthy, knowledge intensive conversational systems. However, existing Retrieval Augmented Generation (RAG) methods typically rely on a retrieval-as-context paradigm that linearizes structured subgraphs into unstructure...
#
KDD2026
BRIDGE: Block-Wise Speculative Coordination for Cloud-Edge Retrieval-Augmented Generation
https://doi.org/10.1145/3770855.3818064
[ "Yuting Li", "Shaoyuan Huang", "Xiangqi Liu", "Yunfeng Zhao", "Xiaofei Wang" ]
Retrieval-augmented generation (RAG) improves factuality by conditioning LLMs on retrieved evidence, yet real-world knowledge is often split across tiers: cloud-based RAG can exploit large public corpora, whereas edge-based RAG is the natural place to access private, user-specific stores. This raises a key question: ho...
#
KDD2026
SkillTracer: Structural Failure Attribution and Refinement of Agentic Skills in Long-Horizon Web Tasks
https://doi.org/10.1145/3770855.3817981
[ "Yuyang Li", "Yiran Dou", "Jie-Jing Shao", "Yueming Lyu", "Ivor Tsang", "Haiyan Yin" ]
Long-horizon web agents frequently fail without knowing where or why execution broke down. This issue is particularly pronounced in skill-based agentic web systems, where failures arise within composite skills whose internal decision processes are not directly traceable, making precise diagnosis and repair especially d...
#
KDD2026
UrbanExpert: Task-Conditioned Multi-Modal Fusion via Semantic Expert Routing for Urban Socioeconomic Prediction
https://doi.org/10.1145/3770855.3818168
[ "Zechen Li", "Hongwei Jia", "Weiming Huang", "Kai Zhao", "Meng Chen" ]
Predicting urban indicators from multi-modal sensing data requires fusing heterogeneous modalities, yet different prediction tasks rely on different feature combinations while semantically related tasks share common cues. Existing approaches either train task-specific models that cannot share knowledge across tasks, or...
#
KDD2026
End-to-End Structured Information Extraction from Mixed-Script Documents in Open Compositional Symbol Systems
https://doi.org/10.1145/3770855.3817959
[ "Zehan Li", "Fu Zhang", "Zhijun Liu", "Jingwei Cheng" ]
Structurally extracting information from mixed-script documents that interleave standard text with open, compositional symbol systems is challenging for both optical character recognition (OCR) and vision–language models (VLMs). This difficulty is epitomized by Jianzi Pu—the ancient Guqin tablature. Unlike closed-set s...
#
KDD2026
TiWeaver: Unified Temporal Dynamics Modeling via Contextual Patching
https://doi.org/10.1145/3770855.3817748
[ "Zhe Li", "Jindong Tian", "Hao Miao", "Zhi Lei", "Chenjuan Guo", "Bin Yang" ]
Multivariate time series forecasting plays a critical role in real-world applications, including weather prediction, stock analysis, and health monitoring. Due to the diversity of data sources, time series exhibit diverse temporal dynamics, often accompanied by various irregularities such as missing values and non-unif...
#
KDD2026
Local Rules, Global Efficiency: Emergent Coordination on Large-Scale Road Networks
https://doi.org/10.1145/3770855.3818071
[ "Zhenhui Jessie Li" ]
Urban transportation systems face a fundamental efficiency paradox: individually optimal routing decisions can create system-wide congestion that increases travel time for all participants. Traffic engineers address this through rigid infrastructure solutions—for example, roundabouts that enforce circular flow, or turn...
#
KDD2026
Mag-Mamba: Modeling Coupled Spatio-temporal Asymmetry for POI Recommendation
https://doi.org/10.1145/3770855.3817697
[ "Zhuoxuan Li", "Tangwei Ye", "Jieyuan Pei", "Haina Liang", "Zhongyuan Lai", "Zihan Liu", "Yiming Wu", "Qi Zhang", "Liang Hu" ]
Next Point-of-Interest (POI) recommendation is a critical task in location-based services, yet it faces the fundamental challenge of coupled spatiotemporal asymmetry inherent in urban mobility. Specifically, transition intents between locations exhibit high asymmetry and are dynamically conditioned on time. Existing me...
#
KDD2026
Local Clustering on Complex Graphs and Complex Hypergraphs
https://doi.org/10.1145/3770855.3818092
[ "Zihao Li", "Dongqi Fu", "Hengyu Liu", "Jingrui He" ]
Local/seeded clustering aims to find a compact cluster near the given starting instances. While most existing studies on graph clustering assume a discrete graph setting (i.e., unweighted, undirected graphs without self-loops), real-world graphs can be more complex. In this paper, we extend the classic non-approximatin...
https://github.com/iDEA-iSAIL-Lab-UIUC/HyperACL
KDD2026
CTR-Sink: Attention Sink for Language Models in Click-Through Rate Prediction
https://doi.org/10.1145/3770855.3817646
[ "Zixuan Li", "Binzong Geng", "Jing Xiong", "Yong He", "Yuxuan Hu", "Jian Chen", "Dingwei Chen", "Xiyu Chang", "Ngai Wong", "Liang Zhang", "Linjian Mo", "Chengming Li", "Chuan Yuan", "Zhenan Sun" ]
Click-Through Rate (CTR) prediction, a core task in recommendation systems, estimates user click likelihood using historical behavioral data. Modeling user behavior sequences as text to leverage Language Models (LMs) for this task has gained traction, owing to LMs' strong semantic understanding and contextual modeling ...
#
KDD2026
Disentangling Multi-View Scanning in Mamba for Network Traffic Anomaly Detection
https://doi.org/10.1145/3770855.3817980
[ "Xinglin Lian", "Chengtai Cao", "Ting Zhong", "Fan Zhou" ]
Network Traffic Anomaly Detection (NTAD) is a critical task in cybersecurity, yet timely and accurate anomaly detection remains challenging. Mamba has emerged as a particularly promising backbone for NTAD due to its linear-time complexity for long-sequence modeling. It further incorporates a dedicated multi-view scanni...
https://github.com/ikun0124/DisenMamba
KDD2026
Unified Multi-Domain Graph Pre-training for Homogeneous and Heterogeneous Graphs via Domain-Specific Expert Encoding
https://doi.org/10.1145/3770855.3817734
[ "Chundong Liang", "Yongqi Huang", "Dongxiao He", "Peiyuan Li", "Yawen Li", "Di Jin", "Weixiong Zhang" ]
Graph pre-training has achieved remarkable success in recent years, delivering transferable representations for downstream adaptation. However, most existing methods are designed for either homogeneous or heterogeneous graphs, thereby hindering unified graph modeling across diverse graph types. This separation contradi...
#
KDD2026
Cognitive Bifurcation: Dual-Progressive Causal Diffusion with Hippocampal Memory for Continual Graph Learning
https://doi.org/10.1145/3770855.3818129
[ "Jiahao Liang", "Carl Yang", "Haoran Yang", "Zhiwen Yu", "Mengzhu Wang", "Kaixiang Yang" ]
Continual Graph Learning (CGL) on non-stationary streams faces the fundamental challenge of adapting to complex distribution shifts, where the entanglement of invariant causal structures and transient environmental noise inevitably leads to catastrophic forgetting. Under such non-stationary conditions, existing methods...
#
KDD2026
Efficient Simple Temporal Cycle Enumeration on Large Graphs with Lightweight Preprocessing
https://doi.org/10.1145/3770855.3817777
[ "Qi Liang", "Dian Ouyang", "Kang Chen", "Fan Zhang", "Xuemin Lin" ]
Temporal cycles are fundamental patterns in graphs, with important applications in finance, security, and neuroscience. In this work, we study the Simple Temporal Cycle Enumeration (STCE) problem, which aims to enumerate all simple cycles with strictly increasing timestamps within a given time window. However, existing...
#
KDD2026
Learning in the Right Subspace: Personalized Differential Private Federated Learning with Noise Filtering
https://doi.org/10.1145/3770855.3817921
[ "Tianchi Liao", "Xiaojun Deng", "Lele Fu", "Sheng Huang", "Bowen Deng", "Hong-Ning Dai", "Chuan Chen" ]
Differential privacy (DP) mechanisms have been widely adopted in federated learning (FL) to enhance model security. However, existing approaches predominantly employ uniform privacy budgets, neglecting personalized requirements arising from heterogeneous user privacy preferences. Such uniform privacy configurations typ...
#
KDD2026
Personalizing Large Language Models with User Profile Memory
https://doi.org/10.1145/3770855.3817930
[ "Yangxu Liao", "Yongheng Deng", "Tianyuan Jiang", "Ju Ren" ]
Large language models (LLMs) are increasingly used in personalized applications, where responses must align with individual user preferences, histories, and profiles. A common approach is to inject user information into the prompt at inference time. However, existing methods typically rely on flat profile representatio...
#
KDD2026
TuneAgent: Agentic Operating System Kernel Tuning with Reinforcement Learning
https://doi.org/10.1145/3770855.3817987
[ "Hongyu Lin", "Yuchen Li", "Haoran Luo", "Zhenghong Lin", "Libo Zhang", "Mingjie Xing", "Yanjun Wu" ]
Linux kernel tuning is essential for optimizing operating system (OS) performance, yet remains challenging due to the complex kernel space, sparse performance feedback, and strong workload sensitivity. We present TuneAgent, an agentic Linux kernel tuning framework powered by rule-based reinforcement learning (RL). Tune...
#
KDD2026
Simple yet Effective Diffusion-based Graph Data Augmentation via Complementary Diffusion Transfer
https://doi.org/10.1145/3770855.3817992
[ "Longlong Lin", "Youan Zhang", "Zeli Wang", "Xin Luo" ]
Graph Neural Networks (GNNs) have emerged as powerful tools for modeling relational data, demonstrating impressive performance across many downstream tasks. Unfortunately, their effectiveness is often hindered by inherent graph imperfections, such as noise and incompleteness. To enhance the robustness of GNNs, numerous...
#
KDD2026
Decoding Intent from Fragments: Structure-Aware Late Interaction for Concept-Level EEG-to-Text Retrieval
https://doi.org/10.1145/3770855.3818019
[ "Rui Lin", "Bo Xu", "Quanhao Zhu", "Boling Zhu", "Chenyuan Wang", "Liang Zhao" ]
While retrieval-based paradigms have emerged as a more viable alternative to generative approaches for EEG-to-Text decoding, existing methods typically compress continuous EEG signals into holistic vectors. However, real-world brain-computer interaction is fundamentally driven by discrete conceptual units, and current ...
#
KDD2026
HyperEdit: Mitigating Hallucinations of Large Language Models via Hyperbolic Representation Editing
https://doi.org/10.1145/3770855.3818153
[ "Tongxu Lin", "Junping Du", "Zhe Xue", "Meiyu Liang", "Runqing Tang" ]
Large language models (LLMs) often generate hallucinations, undermining the reliability of their outputs. While prior work improves truthfulness by contrastive decoding or representation editing, these approaches overlook the hierarchical relationship between truthful and untruthful content, thereby constraining the LL...
#
KDD2026
Retrv-MoE: Scaling Unified Multimodal Retrieval with Sparse Mixture-of-Experts
https://doi.org/10.1145/3770855.3817958
[ "Tongxu Lin", "Jiayin Xiao" ]
Unified multimodal retrieval aims to build a single system capable of handling diverse modalities, tasks, and domains. While recent approaches leveraging multimodal large language models (MLLMs) have shown promise, they face a fundamental dilemma between capacity and interference: scaling dense models improves semantic...
#
KDD2026
Treatment Effect Estimation with Differentiated Networked Effect on Graph Data
https://doi.org/10.1145/3770855.3817678
[ "Xiaofeng Lin", "Han Bao", "Hisashi Kashima" ]
Estimating individual treatment effect (ITE) from observational graph data is crucial for decision-making in the fields such as commerce and medicine. This task is challenging due to interference, where individual outcomes can be influenced by the treatments and covariates of their neighbors. Existing methods attempt t...
#
KDD2026
AbstainGNN: Teaching Graph Neural Networks to Abstain for Graph Classification
https://doi.org/10.1145/3770855.3817804
[ "Xixun Lin", "Zhiheng Zhou", "Zhengyin Zhang", "Yancheng Chen", "Shuai Zhang", "Ge Zhang", "Shichao Zhu", "Lixin Zou", "Chuan Zhou", "Peng Zhang", "Shirui Pan", "Yanan Cao" ]
Graph classification is a core task in graph data mining with widespread real-world applications. Recent advances in graph neural networks (GNNs) have led to substantial performance improvements for graph classification. However, existing GNNs are typically forced to make predictions even under high uncertainty or unkn...
#
KDD2026
Intuition-Guided Latent Reasoning for LLM-Based Recommendation
https://doi.org/10.1145/3770855.3817855
[ "Chang Liu", "Yimeng Bai", "Xiaoyan Zhao", "Yang Zhang", "Qifan Wang", "Fuli Feng", "Wenge Rong" ]
Large Language Models (LLMs) have demonstrated impressive reasoning capabilities in complex problem-solving tasks, motivating their use for preference reasoning in recommender systems. Latent reasoning, which operates in continuous hidden spaces rather than discrete tokens, has recently emerged as a promising paradigm ...
https://github.com/Ten-Mao/IntuRec
KDD2026
CDRRM: Contrast-Driven Rubric Generation for Reliable and Interpretable Reward Modeling
https://doi.org/10.1145/3770855.3817703
[ "Dengcan Liu", "Fengkai Yang", "Xiaohan Wang", "Shurui Yan", "Jiajun Chai", "Jiahao Li", "Yikun Ban", "Zhendong Mao", "Wei Lin", "Guojun Yin" ]
Reward modeling is essential for aligning Large Language Models(LLMs) with human preferences, yet conventional reward models suffer from poor interpretability and heavy reliance on costly expert annotations. While recent rubric-based approaches enhance evaluation transparency, they lack systematic quality control, yiel...
#
KDD2026
Collaborative Memory Augmentation for Generative Recommendation
https://doi.org/10.1145/3770855.3818179
[ "Enze Liu", "Zhen Tian", "Wayne Xin Zhao" ]
Generative Recommendation (GR) has exhibited great potential by modeling item transitions as a sequence-to-sequence task. Despite the success of GR, existing frameworks primarily focus on modeling individual user sequences within a constrained internal parametric space, failing to explicitly leverage cross-user collabo...
#
KDD2026
Lethe: Plasticity-aware Active Forgetting for Resource-Efficient On-Device Continual Learning
https://doi.org/10.1145/3770855.3817621
[ "Haibo Liu", "Chenxin Mao", "Zhenzhe Zheng", "Fan Wu", "Guihai Chen", "He Huang" ]
On-device continual learning (CL) is becoming a critical paradigm for intelligent agents to adapt to evolving task streams in dynamic edge environments. However, existing CL methods primarily focus on optimizing learning performance, while neglecting the fundamental tension between ever-growing accumulated knowledge an...
#
KDD2026
Integrated Learnware Identification and Reuse via Reusability-Aware Metric Learning
https://doi.org/10.1145/3770855.3818210
[ "Hai-Tian Liu", "Peng Tan", "Jian-Dong Liu", "Zhi-Hao Tan", "Zhi-Hua Zhou" ]
The learnware paradigm aims to help users solve machine learning tasks by reusing existing well-trained models instead of building from scratch. These models are accommodated in a learnware dock system, where each learnware consists of a model and a specification that characterizes the model's capability, enabling the ...
#
KDD2026
Mining Point-of-No-Return Boundaries in Constrained Dynamical Systems via Counterfactual Auditing
https://doi.org/10.1145/3770855.3818127
[ "Jia Liu", "Jiaxin Luo", "Lejun Ai", "Yue Wang", "Enpeng Lan", "Yulong Li", "Zongyu Li", "Fuxin Zhang", "Siyuan Liu", "Long Hu", "Yixue Hao", "Min Chen" ]
Safety-critical failures in constrained dynamical environments are often detected only after a violation occurs, while outcome metrics (e.g., success rate, time-to-failure) conflate structural inevitability with decision-induced errors. We formulate failure diagnosis as recoverability boundary discovery and define the ...
#
KDD2026
Distribution-Aware End-to-End Embedding for Streaming Numerical Features in Click-Through Rate Prediction
https://doi.org/10.1145/3770855.3818111
[ "Jiahao Liu", "Hongji Ruan", "Weimin Zhang", "Ziye Tong", "Derick Tang", "Zhanpeng Zeng", "Qinsong Zeng", "Peng Zhang", "Tun Lu", "Ning Gu" ]
This paper explores effective numerical feature embedding for Click-Through Rate prediction in streaming environments. Conventional static binning methods rely on offline statistics of numerical distributions; however, this inherently two-stage process often triggers semantic drift during bin boundary updates. While ne...
#
KDD2026
Constructive Specification for Plug-and-Play Learnware Agents
https://doi.org/10.1145/3770855.3817996
[ "Jian-Dong Liu", "Zi-Chen Zhao", "Hao Sun", "Lin-Xing Wu", "Huan Zhang", "Pengyuan Wang", "Ming Zhao", "Xinyu Chu", "Shu Yan", "Yongbei Zhu", "Weijun Zhong", "Zhi-Hao Tan", "Jing Shang", "Yang Yu", "Zhi-Hua Zhou" ]
Large language models are increasingly deployed at scale as API-accessible, tool-augmented agents, forming a heterogeneous, fast-evolving agent ecosystem. A central challenge is query-level identification: selecting the most suitable agent per query from candidates provided as black-box services, where costly input-out...
#
KDD2026
Understanding Generative Recommendation with Semantic IDs from a Model-scaling View
https://doi.org/10.1145/3770855.3817976
[ "Jingzhe Liu", "Liam Collins", "Jiliang Tang", "Tong Zhao", "Neil Shah", "Mingxuan Ju" ]
Recent advancements in generative models have allowed the emergence of a promising paradigm for recommender systems (RS), known as Generative Recommendation (GR), which tries to unify rich item semantics and collaborative filtering signals. One popular GR paradigm is to use semantic IDs (SIDs), which are discrete codes...
#
KDD2026
A Practical Upper Bound on Selection Bias Effects in Medical Prediction Models
https://doi.org/10.1145/3770855.3818112
[ "Kara Liu", "Maggie Wang", "Russ B. Altman" ]
Selection bias is a common and often unavoidable aspect of real-world data that challenges the generalizability of machine learning models. When models trained on biased data are deployed in the broader target population, poor model generalization may lead to real harm, particularly in high-risk settings such as health...
https://github.com/kara-liu/selection_gap_est/
KDD2026
Dripper: Token-Efficient Main HTML Extraction with a Lightweight LM
https://doi.org/10.1145/3770855.3817915
[ "Mengjie Liu", "Jiahui Peng", "Wenchang Ning", "Pei Chu", "Jiantao Qiu", "Ren Ma", "He Zhu", "Rui Min", "Lindong Lu", "Linfeng Hou", "Kaiwen Liu", "Yuan Qu", "Zhenxiang Li", "Chao Xu", "Zhongying Tu", "Wentao Zhang", "Conghui He" ]
High-quality main content extraction from web pages is a critical prerequisite for constructing large-scale training corpora. While traditional heuristic extractors are efficient, they lack the semantic reasoning required to handle the structural heterogeneity of the modern web. Conversely, well-pretrained generative L...
#
KDD2026
Training-free Truthfulness Detection via Sparse MLP Value Vectors
https://doi.org/10.1145/3770855.3818113
[ "Runheng Liu", "Heyan Huang", "Xingchen Xiao", "Yanghao Zhou", "Zhijing Wu" ]
Large language models (LLMs) are prone to generating factually incorrect content, motivating methods for assessing truthfulness from internal model signals. While supervised probing approaches can be effective, they require labeled data and classifier training. Recent training-free methods avoid parameter optimization ...
#
KDD2026
SetLLM: Set Large Language Model for Cold-Start Item Recommendation
https://doi.org/10.1145/3770855.3817707
[ "Ruochen Liu", "Hao Chen", "Yuanchen Bei", "Lijia Chen", "Qijie Shen", "Feiran Huang", "Fakhri Karray", "Senzhang Wang" ]
Cold-start item recommendation remains a longstanding challenge, as cold items usually lack sufficient interactions to train the behavioral embeddings. Recently, large language model (LLM)-based item cold-start models have tried to address this issue by generating synthetic user-item interactions to train the behaviora...
#
KDD2026
AgentsKG: A Hierarchical Multi-Agent Framework for Open-Domain Knowledge Graph Construction
https://doi.org/10.1145/3770855.3818058
[ "Shilong Liu", "Yongqiang Liu", "Jiye Liu", "Xuan Guo", "Tianpeng Li", "Qiyao Peng", "Wenjun Wang" ]
Knowledge Graph Construction (KGC) is essential for transforming unstructured text into structured knowledge representations. Despite advances in Large Language Models, existing methods treat KGC as a single-pass generation task, conflating extraction, normalization, and validation within a single forward pass. This le...
#
KDD2026
ProfiliTable: Profiling-Driven Tabular Data Processing via Agentic Workflows
https://doi.org/10.1145/3770855.3817982
[ "Wei Liu", "Yang Gu", "Xi Yan", "Zihan Nan", "Beicheng Xu", "Keyao Ding", "Bin Cui", "Wentao Zhang" ]
Table processing—including cleaning, transformation, augmentation, and matching—is a foundational yet error-prone stage in real-world data pipelines. While recent LLM-based approaches show promise for automating such tasks, they often struggle in practice due to ambiguous instructions, complex task structures, and the ...
#
KDD2026
Discovering and Alleviating Data Leakage in Staytime Prediction for Live Streaming Recommendation
https://doi.org/10.1145/3770855.3818187
[ "Weihao Liu", "Xiaopeng Ye", "Chen Zhang", "Haiyuan Zhao", "Xiaoyan Zhao", "Xiao Zhang", "Jun Xu" ]
Staytime prediction plays a central role in live-streaming recommendation. To satisfy the high timeliness requirements of live scenarios, previous methods mainly employ multi-window streaming, splitting each watching record into multiple samples with different temporal horizons. However, our empirical results reveal th...
#
KDD2026
Generalized Range Filtering Approximate Nearest Neighbor Search: Containment and Overlap
https://doi.org/10.1145/3770855.3817986
[ "Yingfan Liu", "Tong Wu", "Jiadong Xie", "Yang Zhao", "Jeffrey Xu Yu", "Jiangtao Cui" ]
Approximate nearest neighbor (ANN) search with range filters has recently garnered significant attention. This paper delves into a generalized form of this problem, i.e., ANN search with exact range-range (RR) predicates on a range-valued attribute, named RR filtering ANN (RRANN). Specifically, given a set of objects c...
https://github.com/FanEDG/MSTG
KDD2026
RouteGoT: Node-Adaptive Routing for Cost-Efficient Graph of Thoughts Reasoning
https://doi.org/10.1145/3770855.3818056
[ "Yuhang Liu", "Ruijie Wang", "Yunlong Chu", "Bing Hao", "Yumeng Lin", "Shengzhong Liu", "Minglai Shao" ]
Large Language Models (LLMs) excel at multi-step reasoning, yet increasing the structural complexity of inference does not consistently improve system-level returns. Methods such as Tree of Thoughts (ToT), Graph of Thoughts (GoT), and Adaptive Graph of Thoughts (AGoT) can boost accuracy on some benchmarks, but often in...
https://github.com/karmaisacat-13/RouteGoT
KDD2026
TRIAD: Ternary Information Routing for Multimodal Time-Series Sensing Signals
https://doi.org/10.1145/3770855.3818194
[ "Yusen Liu", "Shouxing Ma", "Hua Lu", "Xu Cheng", "Xiufeng Liu", "Huan Huo" ]
Multimodal time series captured by heterogeneous sensors are fundamental to real-world applications, yet learning effective representations without labels remains challenging. Existing self-supervised methods adopt a binary paradigm that decomposes information into shared and private components, conflating two distinct...
#
KDD2026
Formalizing and Mitigating Structural Distortion in LLM Attention for Graph Reasoning
https://doi.org/10.1145/3770855.3818097
[ "Donald Loveland", "Puja Trivedi", "Ari Weinstein", "Edward W. Huang", "Danai Koutra" ]
Large Language Models (LLMs) have shown promise for reasoning over Text-Attributed Graphs (TAGs). However, applying LLMs to graphs requires linearizing their structure into sequences, introducing distortion rooted in the graph bandwidth problem. While this distortion has been shown to degrade performance, it is often a...
#
KDD2026
The Price of Fairness in Active Learning: Fundamental Limits and Optimal Label Acquisition
https://doi.org/10.1145/3770855.3818205
[ "Chang Lu", "Yizheng Zhao" ]
When a classifier is trained by actively choosing which examples to label, should its labeling budget depend on which notion of fairness it must satisfy? We answer this with the first information-theoretic lower bounds for group-fair active learning, and the answer is a sharp hierarchy across the three standard notions...
#
KDD2026
SDE : Scale-Difference Evolution Knowledge Distillation
https://doi.org/10.1145/3770855.3817846
[ "Hejie Lu" ]
Logit-based knowledge distillation (KD) is pivotal for efficient model compression and cross-architecture learning. However, conventional methods typically rely on static, single-scale logit alignment, thereby overlooking the semantic evolution trajectory embedded in cross-scale prediction transitions. To bridge this g...
#
KDD2026
LENS-SFL: Learning-Driven Contracts Framework for Personalized Split Federated Learning under Strong Uncertainty
https://doi.org/10.1145/3770855.3817868
[ "Jianfeng Lu", "Yitian Huang", "Yun Xin", "Zhongbo Wu", "Weigang Li", "Guanghui Wen" ]
Split Federated Learning (SFL) enables collaborative training of deep models by partitioning computation across distributed system components while preserving data locality at clients. Existing SFL systems typically rely on static split configurations or assume obedient clients with known capability distributions, whic...
#
KDD2026
PrismFed: Joint Optimization via Dynamic Bayesian Persuasion for Multi-Task Federated Learning under Incomplete Information
https://doi.org/10.1145/3770855.3817867
[ "Jianfeng Lu", "Shicheng Xie", "Yun Xin", "Shuqin Cao", "Weigang Li", "Guanghui Wen" ]
Efficiently allocating limited communication resources in multi-task federated learning is a fundamental challenge. This challenge is particularly pronounced under incomplete information, where the server cannot observe clients' private characteristics and communication resources are scarce and dynamically varying over...
#
KDD2026
CATGX: Causal-Aware Temporal Graph Explanation via Scalable Motif Sampling and Adjustment
https://doi.org/10.1145/3770855.3818100
[ "Mingjian Lu", "Hieu Vu", "Vu K. Le", "Jing Ma", "Yinghui Wu" ]
Interpreting predictions of temporal graph neural networks (TGNNs) is challenging: structural patterns are entangled with temporal dynamics and node-level activity. Existing methods often overemphasize frequent or recent interactions, producing explanations that conflate structural influence with temporal or behavioral...
#
KDD2026
SCALE: Style-Causal Disentanglement with Adaptive Lifelong Expert for Online Latent-domain Anomaly Detection
https://doi.org/10.1145/3770855.3817912
[ "Yimeng Lu", "Yifei Gao", "Tian Lan", "Yingyuan Yang", "Wenjun He", "Meng Wang", "Chen Zhang" ]
Time-series anomaly detection in real-world streams is often challenged by evolving operating conditions, where distribution shifts can be easily mistaken for anomalies. Due to this, we study a new problem, online latent-domain anomaly detection, where domain labels and shift times are unobserved, the number of domains...
#
KDD2026
HierKV: A Coarse-to-Fine Approach with Vision-Aware Banzhaf Values for Multi-Modal KV Cache Compression
https://doi.org/10.1145/3770855.3817663
[ "Zeyi Lu", "Jinpeng Wang", "Yan Feng", "Bin Chen", "Jiawei Li", "Yaowei Wang", "Shu-Tao Xia" ]
Multimodal large language models (MLLMs) are increasingly deployed in Web mining applications such as search, retrieval-augmented assistants, and community question answering, where long and visually rich inputs substantially increase the memory and latency cost of attention through KV caching. Existing KV cache compre...
#
KDD2026
ECHO: Adaptive Community Search over Multimodal Graphs
https://doi.org/10.1145/3770855.3817876
[ "Chengyang Luo", "Zixing Ding", "Qing Liu", "Yifan Zhu", "Yunjun Gao" ]
In this paper, for the first time, we study the community search problem over multimodal graphs. This task aims to identify a query vertex-containing subgraph that is both structurally cohesive and semantically coherent with multimodal query inputs (e.g., text and images). Existing community search methods fail to capt...
#
KDD2026
Decentralized Stochastic Nonconvex Optimization under the (L0, L1)-Smoothness
https://doi.org/10.1145/3770855.3817843
[ "Luo Luo", "Xue Cui", "Tingkai Jia", "Cheng Chen" ]
This paper focuses on the decentralized stochastic optimization problem f(x)=1/m ∑i=1m fi(x) over a connected network of n agents, where each local function has the form of fi(x) = E[F(x;ξ i)] which satisfies the (L0,L1)-smooth condition but possibly nonconvex and each random variable ξi follows distribution Di. We pro...
#
KDD2026
Compress the Easy, Explore the Hard: Difficulty-Aware Entropy Regularization for Efficient LLM Reasoning
https://doi.org/10.1145/3770855.3817726
[ "Qin-Wen Luo", "Sheng Ren", "Xiang Chen", "Rui Liu", "Jun Fang", "Naiqiang Tan", "Sheng-Jun Huang" ]
Chain-of-Thought (CoT) has substantially empowered Large Language Models (LLMs) to tackle complex reasoning tasks, yet the verbose nature of explicit reasoning steps incurs prohibitive inference latency and computational costs, limiting real-world deployment. While existing compression methods—ranging from self-trainin...
#
KDD2026