arXivSub Start free trial

IJCAI 2024 Papers with Code β€” Page 3

International Joint Conference on Artificial Intelligence Β· 296 papers

Optimisation and Approximation in Abstract Argumentation: The Case of Stable Semantics

Matthias Thimm (University of Hagen)

CodeOptimizationComputational EfficiencyBenchmark

🎯 What it does: This paper studies two soft stable semantics (k-stable and k*-stable) and analyzes the complexity and approximation algorithms of the corresponding optimization problems;

OTOcc: Optimal Transport for Occupancy Prediction

Pengteng Li (Shenzhen University), Guang Zhou (Deeproute Inc)

CodeAutonomous DrivingOptimizationComputational EfficiencyPoint Cloud

🎯 What it does: Proposes the OTOcc framework for 3D occupancy prediction by modeling the semantic mapping from pixels to voxels as an Optimal Transport problem;

P2P: Transforming from Point Supervision to Explicit Visual Prompt for Object Detection and Segmentation

Guangqian Guo (Northwestern Polytechnical University), Shan Gao (Northwestern Polytechnical University)

CodeObject DetectionSegmentationTransformerPrompt EngineeringImage

🎯 What it does: Propose a point-supervised object detection and segmentation framework called P2P, which converts point annotations into visual prompts, uses the visual foundation model SAM to generate pseudo-labels, and then re-trains the fully supervised detection/segmentation network with these pseudo-labels.

PACIA: Parameter-Efficient Adapter for Few-Shot Molecular Property Prediction

Shiguang Wu (Tsinghua University), Quanming Yao (Tsinghua University)

CodeMeta LearningDrug DiscoveryGraph Neural NetworkBiomedical Data

🎯 What it does: Proposed PACIA, a parameter-efficient GNN adapter for few-shot molecular property prediction;

ParaILP: A Parallel Local Search Framework for Integer Linear Programming with Cooperative Evolution Mechanism

Peng Lin (Key Laboratory of System Software (Chinese Academy of Sciences) and State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences), Shaowei Cai (Key Laboratory of System Software (Chinese Academy of Sciences) and State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences)

CodeOptimizationBenchmark

🎯 What it does: Proposes ParaILPβ€”a parallel local search framework for general integer linear programming (ILP)β€”that can quickly find high-quality feasible solutions in multi-core environments.

Partial Optimal Transport Based Out-of-Distribution Detection for Open-Set Semi-Supervised Learning

Yilong Ren (University of Science and Technology of China), S. Kevin Zhou (University of Science and Technology of China)

CodeDomain AdaptationAnomaly DetectionOptimizationConvolutional Neural NetworkImage

🎯 What it does: Model OOD detection in open semi-supervised learning as a partial optimal transport problem, design a binary classifier based on quality scores, and jointly train it with existing SSL frameworks (e.g., FixMatch) to achieve end-to-end OOD detection and classification;

PDENNEval: A Comprehensive Evaluation of Neural Network Methods for Solving PDEs

Ping Wei (Sun Yat-sen University), Qingsong Zou (Sun Yat-sen University)

CodeBenchmarkPhysics Related

🎯 What it does: This paper proposes PDENNEval, a unified benchmark platform for evaluating the performance of 12 neural network methods on 19 multidisciplinary PDE problems.

PEACH: Pretrained-Embedding Explanation across Contextual and Hierarchical Structure

Feiqi Cao (University of Sydney), Hyunsuk Chung (FortifyEdge)

CodeClassificationExplainability and InterpretabilityTransformerLarge Language ModelSupervised Fine-TuningText

🎯 What it does: Propose PEACH, which constructs decision trees using context embeddings from any pre-trained language model to provide global and local interpretable hierarchical explanations for text classification.

Personalized Federated Learning for Cross-City Traffic Prediction

Yu Zhang (Shandong University), Lizhen Cui (Shandong University)

CodeFederated LearningSafty and PrivacyRecurrent Neural NetworkGraph Neural NetworkTime Series

🎯 What it does: Proposes a personalized federated learning framework named pFedCTP for cross-city traffic prediction, protecting data privacy of source and target cities.

Personalized Heart Disease Detection via ECG Digital Twin Generation

Yaojun Hu (Zhejiang University), Jian Wu (Zhejiang University)

CodeAnomaly DetectionConvolutional Neural NetworkTransformerGenerative Adversarial NetworkTime SeriesBiomedical DataElectrocardiogram

🎯 What it does: Proposes a proactive learning framework called LAVQ-Editor, which enhances the sensitivity and personalized diagnosis of cardiac disease detection models by generating personalized ECG digital twins.

PHSIC against Random Consistency and Its Application in Causal Inference

Jue Li (Shanxi University), Saixiong Liu (Shanxi University)

CodeTabular

🎯 What it does: Propose Pure HSIC (PHSIC) to eliminate the random consistency of HSIC and apply it to causal inference models.

Physics-Informed Neural Networks: Minimizing Residual Loss with Wide Networks and Effective Activations

Nima Hosseini Dashtbayaz (University of Western Ontario), Charles X. Ling (University of Western Ontario)

CodeOptimizationPhysics Related

🎯 What it does: This paper investigates the properties of residual loss in physics-informed neural networks (PINNs), deriving that wide networks can globally minimize residual loss, and pointing out that the higher-order derivatives of activation functions must be bijective to enhance expressiveness, thereby verifying the effectiveness of periodic activation functions (e.g., sine) in solving first- to second-order PDEs.

Pluggable Watermarking of Deepfake Models for Deepfake Detection

Han Bao (Zhejiang University), Wenzhi Chen (Zhejiang University)

CodeGenerationAnomaly DetectionConvolutional Neural NetworkImage

🎯 What it does: Proposed a pluggable, low-cost active model watermark framework to embed extractable watermarks into the decoders of trained Deepfake generation models, enabling detection of generated images;

Pointsoup: High-Performance and Extremely Low-Decoding-Latency Learned Geometry Codec for Large-Scale Point Cloud Scenes

Kang You (Nanjing University of Aeronautics and Astronautics), Dandan Ding (Hangzhou Normal University)

CodeCompressionGraph Neural NetworkTransformerPoint Cloud

🎯 What it does: Proposed an efficient learning-based geometric encoder-decoder called Pointsoup for compressing and rapidly decoding large-scale point clouds.

Practical Anytime Algorithms for Judicious Partitioning of Active Directory Attack Graphs

Yumeng Zhang (University of Adelaide), Hung Nguyen (University of Adelaide)

CodeOptimizationSafty and PrivacyGraph

🎯 What it does: This paper proposes a new graph theory problemβ€”maximizing the number of disconnected pairs between a source and target node by edge deletion under a budget constraint, and applies it to security reinforcement of the Windows Active Directory (AD) system;

Pre-DyGAE: Pre-training Enhanced Dynamic Graph Autoencoder for Occupational Skill Demand Forecasting

Xi Chen (University of Science and Technology of China), Hui Xiong (Hong Kong University of Science and Technology (Guangzhou))

CodeGraph Neural NetworkAuto EncoderContrastive LearningTextTime SeriesFinance Related

🎯 What it does: Propose a method based on a pre-trained enhanced dynamic graph autoencoder (Pre-DyGAE) for predicting skill demand from a career perspective.

Pre-training General User Representation with Multi-type APP Behaviors

Yuren Zhang (University of Science and Technology of China), Yang Yu (University of Science and Technology of China)

CodeRecommendation SystemTransformerSupervised Fine-TuningAuto EncoderContrastive LearningTabularSequential

🎯 What it does: Proposed a Multi-Type App Usage Fusion Network (MAFN) that obtains general user representations through pre-training and achieves significant improvements on three types of downstream tasks.

Predictive Modeling with Temporal Graphical Representation on Electronic Health Records

Jiayuan Chen (Ohio State University), Ping Zhang (Ohio State University)

CodeClassificationExplainability and InterpretabilityGraph Neural NetworkTransformerBiomedical DataElectronic Health Records

🎯 What it does: Proposed a time-heterogeneous graph-based representation method for electronic health records (EHR) and designed the Temporal Graph Transformer (TRANS) model to simultaneously capture structural information of clinical events and temporal dynamics between follow-ups for diagnostic prediction.

Primal Grammars Driven Automated Induction

Adel Bouhoula (Arabian Gulf University), Miki Hermann (cole Polytechnique)

CodeComputational EfficiencyBenchmark

🎯 What it does: This paper proposes an automated inductive proof method that can detect and capture divergence during the proof process, generating new lemmas using primal grammars to complete the proof;

Probabilistic Contrastive Learning for Domain Adaptation

Junjie Li (Beijing University of Posts and Telecommunications), Man Zhang (Beijing University of Posts and Telecommunications)

CodeDomain AdaptationContrastive LearningImage

🎯 What it does: Proposed a probability-based contrastive learning method to address the issue of feature and class weight deviation in domain adaptation.

Prompt Learning for Generalized Vehicle Routing

Fei Liu (City University of Hong Kong), Mingxuan Yuan (Huawei Noah's Ark Lab)

CodeOptimizationMeta LearningTransformerReinforcement LearningPrompt EngineeringGraph

🎯 What it does: Leveraging prompt learning to achieve rapid zero-shot adaptation of pre-trained NCO models for addressing cross-distribution vehicle routing problems.

Prompt-enhanced Network for Hateful Meme Classification

Junxi Liu (South China Normal University), Fenghuan Li (Guangdong University of Technology)

CodeClassificationRecurrent Neural NetworkTransformerLarge Language ModelPrompt EngineeringContrastive LearningMultimodality

🎯 What it does: Proposes a Prompt-Enhanced Network (Pen) framework that extends prompt learning into the feature space, combining multi-view perception and prompt-aware contrastive learning for hate meme classification.

Purpose Enhanced Reasoning through Iterative Prompting: Uncover Latent Robustness of ChatGPT on Code Comprehension

Yi Wang (North Carolina State University), Xu Liu (North Carolina State University)

CodeAI Code AssistantTransformerLarge Language ModelPrompt EngineeringText

🎯 What it does: Propose and implement a multi-module prompting framework named Perthept, which enhances the robustness and quality of ChatGPT in code comment generation through two iterative steps: Chain-of-Structure and Reasoning-Enhancement.

Putting Back the Stops: Integrating Syntax with Neural Topic Models

Mayank Nagda (RPTU Kaiserslautern-Landau), Sophie Fellenz (RPTU Kaiserslautern-Landau)

CodeRepresentation LearningAuto EncoderText

🎯 What it does: Designed a neural topic model named SyConNTM that can simultaneously learn syntactic and semantic topics without preprocessing, and automatically identify stop words through a context network.

R2V-MIF: Rule-to-Vector Contrastive Learning and Multi-channel Information Fusion for Therapy Recommendation

Nengjun Zhu (Shanghai University), Huanjing Gao (Shanghai University)

CodeRecommendation SystemGraph Neural NetworkContrastive LearningGraphBiomedical Data

🎯 What it does: Proposed the R2V-MIF model, which achieves treatment recommendation through rule-to-vector contrastive learning and multi-channel information fusion.

Real-World Networks Are Low-Dimensional: Theoretical and Practical Assessment

Tobias Friedrich (University of Potsdam), Leon Schiller (ETH)

CodeGraph

🎯 What it does: This paper proves through theoretical analysis and a linear-time algorithm that the hidden geometric dimension of real-world networks is extremely low (1~10 dimensions), and proposes a method to rapidly estimate this dimension on large-scale datasets.

Recall, Retrieve and Reason: Towards Better In-Context Relation Extraction

Guozheng Li (Southeast University), Zijie Xu (Southeast University)

CodeRetrievalKnowledge DistillationRepresentation LearningTransformerLarge Language ModelPrompt EngineeringTextBenchmarkRetrieval-Augmented GenerationChain-of-Thought

🎯 What it does: Propose a RE4 recall-retrieve-reason framework, which first generates entity pairs consistent with the input sentence as queries using open-source LLMs through ontological knowledge, then retrieves corresponding examples from training corpora, and finally performs in-context reasoning on the retrieved examples, significantly improving relation extraction performance.

Reconfigurability-Aware Selection for Contrastive Active Domain Adaptation

Zeyu Zhang (Jilin University), Shaojie Zhang (Jilin University)

CodeClassificationDomain AdaptationContrastive LearningImage

🎯 What it does: Proposes the RASC (Reconfigurability-Aware Selection for Contrastive Active Domain Adaptation) method, combining reconfigurability-aware sample selection with contrastive learning-based progressive domain adaptation, and introduces a one-annotation variant RASC-Ob

Reconstruction Weighting Principal Component Analysis with Fusion Contrastive Learning

Qianqian Wang (Xidian University), Quanxue Gao (Xidian University)

CodeClassificationRepresentation LearningContrastive LearningImageBiomedical Data

🎯 What it does: This paper proposes an unsupervised feature extraction method that combines reconstruction-weighted principal component analysis (Weighted PCA) with contrastive learning;

Redefining Contributions: Shapley-Driven Federated Learning

Nurbek Tastan (Mohamed bin Zayed University of Artificial Intelligence), Karthik Nandakumar (Mohamed bin Zayed University of Artificial Intelligence)

CodeFederated LearningExplainability and InterpretabilityConvolutional Neural NetworkImage

🎯 What it does: This paper studies the use of Shapley values in federated learning to evaluate the contributions of participants, and proposes the ShapFed algorithm based on class-level Shapley values, the weighted aggregation method ShapFed-WA, and contribution-based personalized updates.

Reframing Spatial Reasoning Evaluation in Language Models: A Real-World Simulation Benchmark for Qualitative Reasoning

Fangjun Li (University of Leeds), Anthony G. Cohn (University of Leeds)

CodeData SynthesisTransformerLarge Language ModelTextTabularBenchmark

🎯 What it does: This paper proposes RoomSpace, a spatial reasoning evaluation benchmark based on 3D simulated scenarios, providing diverse object layouts, topological, directional, and distance relationships, and introducing a logical consistency checking tool;

Reinforcement Nash Equilibrium Solver

Xinrun Wang (Nanyang Technological University), Bo An (Nanyang Technological University)

CodeOptimizationGraph Neural NetworkReinforcement LearningGraph

🎯 What it does: This paper proposes a game modification method based on reinforcement learningβ€”REinforcement Nash Equilibrium Solver (RENES), which trains a general strategy to modify normal-form games of any size, then applies existing approximate solvers (such as Ξ±-rank, CE, FP, PRD) to the modified game, ultimately obtaining solutions closer to Nash equilibrium in the original game.

Relevant Irrelevance: Generating Alterfactual Explanations for Image Classifiers

Silvan Mertes (University of Augsburg), Elisabeth AndrΓ© (University of Augsburg)

CodeClassificationExplainability and InterpretabilityGenerative Adversarial NetworkImage

🎯 What it does: This study proposes a GAN-based generative framework to produce Alterfactual (modifiable irrelevant features) and Counterfactual (decision-changing) explanations for black-box image classifiers, which can preserve the original decision while maximizing modifications to irrelevant features.

ReliaAvatar: A Robust Real-Time Avatar Animator with Integrated Motion Prediction

Bo Qian (Xi'an Jiaotong University), Xing Wei (Xi'an Jiaotong University)

CodePose EstimationRecurrent Neural NetworkTransformerTime SeriesSequential

🎯 What it does: This paper proposes a real-time, robust full-body avatar animator called ReliaAvatar, which can continuously generate complete skeleton motions even when wearable device signals are of low quality or missing.

Reschedule Diffusion-based Bokeh Rendering

Shiyue Yan (Tsinghua University), Shaojun Liu

CodeImage TranslationDiffusion modelImage

🎯 What it does: Propose a prior-based sampling and noise scheduling method to improve diffusion models in Bokeh rendering from small aperture images to large aperture images.

Rethinking Centered Kernel Alignment in Knowledge Distillation

Zikai Zhou (East China Normal University), Shaohui Lin (East China Normal University)

CodeClassificationObject DetectionKnowledge DistillationConvolutional Neural NetworkTransformerImage

🎯 What it does: Investigated the theoretical foundations of Centered Kernel Alignment (CKA) in knowledge distillation, and proposed two concise and efficient distillation frameworks: relation-based CKA (RCKA) and patch-based CKA (PCKA), to achieve effective alignment of features and logit layers.

Rethinking the Effectiveness of Graph Classification Datasets in Benchmarks for Assessing GNNs

Zhengdao Li (Guangzhou University), Kai Hwang (Chinese University of Hong Kong)

CodeClassificationData SynthesisGraph Neural NetworkGraphBenchmark

🎯 What it does: Designed and implemented new experimental protocols and metrics to evaluate whether graph classification benchmark datasets can effectively distinguish GNNs from baseline methods, and proposed a new effectiveness metric E and a synthetic data generation framework with controllable correlation.

Rethinking the Soft Conflict Pseudo Boolean Constraint on MaxSAT Local Search Solvers

Jiongzhi Zheng (Huazhong University of Science and Technology), Kun He (Huazhong University of Science and Technology)

CodeOptimizationBenchmark

🎯 What it does: Proposes integrating Soft conflict Pseudo Boolean constraints into the clause weighting system of local search MaxSAT and designs an adaptive weight update strategy.

Revisiting Neural Networks for Continual Learning: An Architectural Perspective

Aojun Lu (Sichuan University), Yanan Sun (Sichuan University)

CodeClassificationNeural Architecture SearchConvolutional Neural NetworkImage

🎯 What it does: Systematically experiment on the impact of network architecture (width, depth, skip connections, global pooling, downsampling, etc.) in continual learning (Task IL and Class IL), discovering that wider, shallower, skip-connection-equipped networks with removed GAP (Task IL) or retained GAP (Class IL) are more favorable. Based on these insights, we propose the ArchCraft NAS method, automatically searching for lightweight networks suitable for continual learning (AlexAC, ResAC).

RoboFusion: Towards Robust Multi-Modal 3D Object Detection via SAM

Ziying Song (Beijing Jiaotong University), Li Wang (Beijing Institute of Technology)

CodeObject DetectionAutonomous DrivingConvolutional Neural NetworkTransformerImageMultimodalityPoint Cloud

🎯 What it does: Proposes the RoboFusion framework, combining the visual foundation model SAM with multi-modal information to enhance the robustness of 3D object detection in noisy scenarios.

Robust Contrastive Multi-view Kernel Clustering

Peng Su (Sichuan University), Jiancheng Lv (Sichuan University)

CodeOptimizationRepresentation LearningContrastive LearningMultimodality

🎯 What it does: This paper proposes a robust contrastive learning multi-view kernel clustering framework named R-CMK, aiming to enhance kernel quality by processing pseudo-negative samples through reverse gradients, thereby improving multi-view clustering performance.

Robust Heterophilic Graph Learning against Label Noise for Anomaly Detection

Junhang Wu (Wuhan University), Yilong Zang (Wuhan University)

CodeAnomaly DetectionGraph Neural NetworkContrastive LearningGraph

🎯 What it does: This paper proposes the NRGL framework, which addresses graph anomaly detection tasks with graph heterogeneity and label noise. It first splits the original graph into homophilic (high homophily) and heterophilic views using an edge discriminator. Then, it obtains robust node representations through unsupervised contrastive learning and enhances the structure via low-pass/high-pass graph filters. Subsequently, it divides the training nodes into clean, confident, and remaining groups based on the small loss criterion and designs a sampler based on degree, class frequency, and confidence for balanced training.

ROCES: Robust Class Expression Synthesis in Description Logics via Iterative Sampling

N'Dah Jean Kouagou (Paderborn University), Axel-Cyrille Ngonga Ngomo (Paderborn University)

CodeData SynthesisTransformerGraph

🎯 What it does: Propose a new class expression learning problem GLP and introduce the ROCES algorithm, achieving efficient class expression reasoning for any number of samples through iterative sampling and neural synthesizers.

ROME: Robust Multi-Modal Density Estimator

Anna MΓ©szΓ‘ros (TU Delft), Jens Kober (TU Delft)

CodeRepresentation LearningMultimodalityTabularSequential

🎯 What it does: Propose a robust multi-modal probability density estimation method called ROME, which is based on clustering, decorrelation, normalization, and then kernel density estimation.

Sample Quality Heterogeneity-aware Federated Causal Discovery through Adaptive Variable Space Selection

Xianjie Guo (Hefei University of Technology), Xiaoxiao Li (The University of British Columbia)

CodeFederated LearningGraphTabularBiomedical DataBenchmark

🎯 What it does: Propose the FedACD method to address the issue of sample quality heterogeneity across different variable spaces in federated causal discovery, using adaptive masking to transmit only the causal relationships learned from high-quality variable spaces.

Sampling Winners in Ranked Choice Voting

Matthew Iceland (University of Rochester), Joseph Saber (University of Rochester)

CodeTabular

🎯 What it does: The study investigates the feasibility of using ranking voting (RCV) with sample voting to predict complete election outcomes, and evaluates its prediction accuracy at both theoretical and empirical levels.

Scalable Mechanism Design for Multi-Agent Path Finding

Paul Friedrich (ETH AI Center), Sven Seuken (ETH AI Center)

CodeOptimizationGraphBenchmark

🎯 What it does: Propose a scalable mechanism design framework in multi-agent path finding (MAPF) to achieve strategyproofness and individual rationality (IR), and design three mechanisms: PCBS, EPBS, and MCPP.

Scaling Up Unbiased Search-based Symbolic Regression

Paul Kahlmeyer (Friedrich Schiller University Jena), Henrik Voigt (Friedrich Schiller University Jena)

CodeOptimizationData-Centric LearningBenchmarkPhysics Related

🎯 What it does: A symbolic regression method based on unbiased expression DAG search is studied, and a variable expansion technique is proposed to reduce the search space

Scene-Adaptive Person Search via Bilateral Modulations

Yimin Jiang (Dalian Maritime University), Yang Wang (Hefei University of Technology)

CodeRetrievalDomain AdaptationConvolutional Neural NetworkTransformerContrastive LearningImageBenchmark

🎯 What it does: Proposed a scene-adaptive person search framework named SEAS, which eliminates scene noise through bilateral modulation and maintains feature consistency of the same person across different scenes;

Score-CDM: Score-Weighted Convolutional Diffusion Model for Multivariate Time Series Imputation

Shunyang Zhang (Central South University), Jian Zhang (Central South University)

CodeData SynthesisConvolutional Neural NetworkTransformerDiffusion modelScore-based ModelTime Series

🎯 What it does: Proposed a diffusion model called Score-CDM based on score-weighted convolution for imputing missing values in multivariate time series.

ScreenAgent: A Vision Language Model-driven Computer Control Agent

Runliang Niu (Jilin University), Qi Wang (Jilin University)

CodeTransformerSupervised Fine-TuningReinforcement LearningAgentic AIVision Language ModelMultimodality

🎯 What it does: Built a computer control environment and agent based on a vision-language model (VLM), enabling mouse and keyboard operations on real computer screens, and trained the ScreenAgent model.

ScreenAI: A Vision-Language Model for UI and Infographics Understanding

Gilles Baechler (Google DeepMind), Abhanshu Sharma (Google DeepMind)

CodeTransformerLarge Language ModelVision Language ModelImageMultimodality

🎯 What it does: Propose ScreenAI, a vision-language model for UI and infographics, achieving cross-domain understanding through self-supervised pre-training and data generated by large-scale LLMs.

SDformer: Transformer with Spectral Filter and Dynamic Attention for Multivariate Time Series Long-term Forecasting

Ziyu Zhou (Beijing University of Technology), Zhen Yang (Beijing University of Technology)

CodeTransformerTime Series

🎯 What it does: Propose the SDformer architecture, addressing the issue of smooth attention distribution in long-range forecasting of multivariate time series through Spectral-Filter-Transform (SFT) and Dynamic-Directional-Attention (DDA).

SeeDRec: Sememe-based Diffusion for Sequential Recommendation

Haokai Ma (Shandong University), Zhanhui Kang (Tencent)

CodeRecommendation SystemPrompt EngineeringDiffusion modelTextSequential

🎯 What it does: Proposed SeeDRec, a diffusion framework based on recommendation semantic units (sememe), to enhance the performance of sequential recommendations.

Self-adaptive Extreme Penalized Loss for Imbalanced Time Series Prediction

Yiyang Wang (Dalian Maritime University), Yuhan Guo (Dalian Maritime University)

CodeAnomaly DetectionRecurrent Neural NetworkTime Series

🎯 What it does: This paper proposes an adaptive extreme penalty loss function (EPL), combining it with Empirical Wavelet Transform decomposition and an LSTM with attention mechanism to form a framework capable of accurately predicting both extreme and normal events in imbalanced time series.

Self-Supervised Learning for Enhancing Spatial Awareness in Free-Hand Sketches

Xin Wang (Shanghai Jiao Tong University), Lei Xu (Shanghai Jiao Tong University)

CodeRestorationRecurrent Neural NetworkGraph Neural NetworkAuto EncoderImage

🎯 What it does: Propose a self-supervised task called 'Sketch Reorganization' to correct stroke position errors in hand-drawn sketches, and design the SketchGloc model to utilize multi-graph structures to learn spatial layout information, enhancing the robustness of sketch representations.

SemanticMask: A Contrastive View Design for Anomaly Detection in Tabular Data

Shuting Tao (Zhejiang University), Xiangming Meng (Zhejiang University)

CodeAnomaly DetectionLarge Language ModelContrastive LearningTabular

🎯 What it does: Proposes a contrastive learning data augmentation method called SemanticMask based on semantic information for unsupervised anomaly detection in tabular data.

SEMANTIFY: Unveiling Memes with Robust Interpretability beyond Input Attribution

Dibyanayan Bandyopadhyay (Indian Institute of Technology Patna), Asif Ekbal (Indian Institute of Technology Jodhpur)

CodeClassificationExplainability and InterpretabilityTransformerLarge Language ModelVision Language ModelMultimodality

🎯 What it does: Propose the SEMANTIFY framework, which employs a multi-step filtering mechanism combining GPT-2 and CLIP to generate interpretable keywords for determining the offensiveness of online memes.

SEMv3: A Fast and Robust Approach to Table Separation Line Detection

Chunxia Qin (University of Science and Technology of China), Jun Du (University of Science and Technology of China)

CodeRecognitionSegmentationConvolutional Neural NetworkTransformerTabular

🎯 What it does: This paper proposes the SEMv3 framework, achieving fast and robust table structure recognition under the split-merge paradigm.

Sentence-Level or Token-Level? A Comprehensive Study on Knowledge Distillation

Jingxuan Wei (Shenyang Institute of Computing Technology Chinese Academy of Sciences), Ruifeng Guo (Shenyang Institute of Computing Technology Chinese Academy of Sciences)

CodeKnowledge DistillationTransformerLarge Language ModelText

🎯 What it does: This paper systematically compares the performance of sentence-level and word-level knowledge distillation in neural machine translation, and proposes a dynamic gated hybrid distillation method.

SGDCL: Semantic-Guided Dynamic Correlation Learning for Explainable Autonomous Driving

Chengtai Cao (City University of Hong Kong), Yung-Hui Li (Foxconn Research)

CodeAutonomous DrivingExplainability and InterpretabilityConvolutional Neural NetworkGraph Neural NetworkVision Language ModelContrastive LearningImageTextMultimodality

🎯 What it does: This paper proposes a model called SGDCL, which combines semantic-guided category feature learning and dynamic association learning to achieve interpretable autonomous driving behavior and explanation prediction.

Shadow-Free Membership Inference Attacks: Recommender Systems Are More Vulnerable Than You Thought

Xiaoxiao Chi (Macquarie University), Hongsheng Hu (CSIRO Data61)

CodeRecommendation SystemAdversarial AttackTabularBenchmark

🎯 What it does: Propose a shadow-free membership inference attack that directly utilizes personalized recommendations from a recommendation system based on users' historical interactions to determine membership status.

Simple Contrastive Multi-View Clustering with Data-Level Fusion

Caixuan Luo (Guangxi Normal University), Xiaofeng Zhu (Guangxi Normal University)

CodeRepresentation LearningContrastive LearningMultimodality

🎯 What it does: Proposed a simple contrastive multi-view clustering framework SCM based on data-level fusion;

SketchEdit: Editing Freehand Sketches at the Stroke-Level

Tengjie Li (Shanghai Jiao Tong University), Lei Xu (Shanghai Jiao Tong University)

CodeGenerationTransformerDiffusion modelAuto EncoderImage

🎯 What it does: Proposes SketchEdit, a stroke-level editing framework for freehand sketches based on diffusion models, which can decouple strokes and generate edited results in reasonable positions.

Spatial-Temporal Perceiving: Deciphering User Hierarchical Intent in Session-Based Recommendation

Xiao Wang (University of Electronic Science and Technology of China), Shuang Liang (University of Electronic Science and Technology of China)

CodeRecommendation SystemGraph Neural NetworkContrastive LearningSequential

🎯 What it does: This paper proposes a hierarchical intent-based conversational recommendation framework called HearInt, aiming to simultaneously decouple and model user behavior from both the temporal aspect (long-term and short-term intents) and the spatial aspect (small-scale and large-scale intents).

Spear: Evaluate the Adversarial Robustness of Compressed Neural Models

Chong Yu (Fudan University), Jiayuan Fan (Fudan University)

CodeClassificationObject DetectionCompressionKnowledge DistillationAdversarial AttackConvolutional Neural NetworkTransformerImage

🎯 What it does: Designed and implemented a specialized adversarial attack method called Spear for sparse and quantized compressed models to evaluate the adversarial robustness of compressed models.

SpecAR-Net: Spectrogram Analysis and Representation Network for Time Series

Yi Dong (Intelligent Science & Technology Academy of CASIC), Zhe Ma (Intelligent Science & Technology Academy of CASIC)

CodeClassificationAnomaly DetectionConvolutional Neural NetworkRecurrent Neural NetworkTransformerTime Series

🎯 What it does: Propose SpecAR-Net, which utilizes time-frequency joint analysis and representation networks to extract global trends, periodicity, and mutation features from time series, achieving a unified multi-task model.

Speech-Forensics: Towards Comprehensive Synthetic Speech Dataset Establishment and Analysis

Zhoulin Ji (Xi'an Jiaotong University), Chao Shen (Xi'an Jiaotong University)

CodeRecognitionAnomaly DetectionRecurrent Neural NetworkTransformerBenchmarkAudio

🎯 What it does: Propose the Speech-Forensics dataset and design the TEST model to achieve authenticity determination of synthetic speech, multiple forged segment localization, and synthesis algorithm identification;

SPGNet: A Shape-prior Guided Network for Medical Image Segmentation

Zhengxuan Song (Sun Yat-sen University), Kun Zeng (Sun Yat-sen University)

CodeSegmentationConvolutional Neural NetworkTransformerImageBiomedical DataUltrasound

🎯 What it does: Developed SPGNet, a dual-path collaborative network integrating statistical shape models to guide medical image segmentation.

Stochastic Neural Simulator for Generalizing Dynamical Systems across Environments

Liu Jiaqi (Jilin University), Bo Yang (Jilin University)

CodeDomain AdaptationTime SeriesPhysics RelatedStochastic Differential EquationOrdinary Differential Equation

🎯 What it does: Proposes a stochastic neural ODE process model called CoNDP for learning and predicting dynamic systems from sparse observations under different environments, capable of capturing environmental commonalities and differences while rapidly adapting to new environments.

SVD-AE: Simple Autoencoders for Collaborative Filtering

Seoyoung Hong (Yonsei University), Noseong Park (Korea Advanced Institute of Science and Technology)

CodeRecommendation SystemAuto EncoderTabular

🎯 What it does: Designed a linear autoencoder called SVD-AE based on truncated singular value decomposition (SVD) for closed-form solutions in collaborative filtering, achieving improved accuracy, efficiency, and noise robustness without iterative training.

Symplectic Neural Gaussian Processes for Meta-learning Hamiltonian Dynamics

Tomoharu Iwata (NTT Corporation), Yusuke Tanaka (NTT Corporation)

CodeMeta LearningRecurrent Neural NetworkTime SeriesSequentialPhysics Related

🎯 What it does: This paper proposes a meta-learning based Hamiltonian dynamics model that can quickly adapt to the dynamics of new systems with only a small amount of data.

TAI++: Text as Image for Multi-Label Image Classification by Co-Learning Transferable Prompt

Xiangyu Wu (Nanjing University of Science and Technology), Jianfeng Lu (Nanjing University of Science and Technology)

CodeClassificationTransformerPrompt EngineeringVision Language ModelContrastive LearningImage

🎯 What it does: Proposes a pseudo visual prompt (PVP) module and a transferable prompt co-learning strategy based on pre-trained vision-language models to achieve multi-label image classification without relying on large-scale annotated image data;

Temporal Domain Generalization via Learning Instance-level Evolving Patterns

Yujie Jin (Peking University), Liantao Ma (Peking University)

CodeData SynthesisDomain AdaptationTabularTime SeriesStochastic Differential EquationOrdinary Differential Equation

🎯 What it does: Propose the CTOT framework, which generates temporal trajectories using instance-level optimal transport and predicts future domains through neural differential equations, enhancing Temporal Domain Generalization performance.

The Orthogonality of Weight Vectors: The Key Characteristics of Normalization and Residual Connections

Zhixing Lu (Dalian University of Technology), Hongfei Lin (Dalian University of Technology)

CodeRepresentation LearningConvolutional Neural NetworkTransformerImageText

🎯 What it does: Studied how combining normalization and residual connections in deep neural networks enhances the orthogonality of weight vectors, thereby improving feature learning capabilities.

The Trembling-Hand Problem for LTLf Planning

Pian Yu (University of Oxford), Moshe Vardi (Rice University)

CodeOptimizationRobotic IntelligenceReinforcement LearningTabular

🎯 What it does: This paper addresses the errors caused by agents making 'trembling-hand' mistakes during the execution of temporally extended goals (LTLf), studying how to generate strategies that maximize the probability of goal satisfaction in deterministic and non-deterministic planning domains.

TIM: An Efficient Temporal Interaction Module for Spiking Transformer

Sicheng Shen (Chinese Academy of Sciences), Yi Zeng (Chinese Academy of Sciences)

CodeClassificationRecognitionSpiking Neural NetworkTransformerVideoTime Series

🎯 What it does: Proposes a plug-and-play time interaction module (TIM) to enhance the temporal information utilization efficiency of spiking transformers when processing time series data.

Toward a Manifold-Preserving Temporal Graph Network in Hyperbolic Space

Viet Quan Le (VNU University of Engineering and Technology), Viet Cuong Ta (VNU University of Engineering and Technology)

CodeRepresentation LearningGraph Neural NetworkGraphTime Series

🎯 What it does: Proposed a time series graph network (HMPTGN) that directly operates on a hypersphere, learning spatial and temporal relationships in dynamic graphs by preserving the hypersurface structure.

Towards a Framework for Learning of Algorithms: The Case of Learned Comparison Sorts

Philipp Kunz (University of Stuttgart), Marco Aiello (University of Stuttgart)

CodeClassificationOptimizationRepresentation LearningHyperparameter SearchTabular

🎯 What it does: Proposed a generic algorithm learning framework that models algorithms as learnable transfer operators using computational graphs and isomorphic algebra, with comparison sorting algorithms as a case study for learning and evaluation.

Towards a Pretrained Model for Restless Bandits via Multi-arm Generalization

Yunfan Zhao (Harvard University), Milind Tambe (Harvard University)

CodeReinforcement Learning

🎯 What it does: This paper develops a pre-trained RMAB model called PreFeRMAB, which supports multi-arm generalization, streaming arms joining/leaving, and nonlinear rewards in continuous states.

Towards Automatic Composition of ASP Programs from Natural Language Specifications

Manuel Borroto Santana (University of Calabria), Francesco Ricca (University of Calabria)

CodeAI Code AssistantTransformerText

🎯 What it does: This paper proposes a dataset named NL2CNL and a tool called NL2ASP, implementing a two-step process for automatically generating ASP programs from natural language descriptions;

Towards Exact Computation of Inductive Bias

Akhilan Boopathy (Massachusetts Institute of Technology), Ila Fiete (Massachusetts Institute of Technology)

CodeImageBenchmark

🎯 What it does: Propose a practical method to directly estimate the inductive bias required for a given task: sample random hypotheses from the hypothesis space, estimate their error distribution on the test set, and calculate the negative logarithm of the probability of achieving a specified error threshold.

Towards Generalizable Neural Solvers for Vehicle Routing Problems via Ensemble with Transferrable Local Policy

Chengrui Gao (Nanjing University), Chao Qian (Nanjing University)

CodeOptimizationTransformerReinforcement LearningGraphBenchmark

🎯 What it does: Propose a neural solver that integrates global and local strategies for the vehicle routing problem.

Trusted Multi-view Learning with Label Noise

Cai Xu (Xidian University), Wei Zhao (Xidian University)

CodeClassificationRepresentation LearningImageTextTabular

🎯 What it does: Propose the Trusted Multi-view Noise Refining (TMNR) method, which utilizes evidence theory to generate beliefs and uncertainties at each view, designs view-specific noise association matrices, transforms original opinions into 'noise opinions' consistent with noisy labels, aggregates multi-view opinions via Dempster's rule, and trains the model using generalized maximum likelihood loss. Additionally, uncertainty-guided regularization and view consistency constraints are employed to enhance robustness against label noise.

Unified Single-Stage Transformer Network for Efficient RGB-T Tracking

Jianqiang Xia (Intelligent Game and Decision Lab), Jian Zhao (China Telecom)

CodeObject TrackingTransformerMultimodality

🎯 What it does: Designed a unified single-stage Transformer network called USTrack, integrating feature extraction, fusion, and relationship modeling for RGB-T tracking.

Unified Unsupervised Salient Object Detection via Knowledge Transfer

Yao Yuan (Nanjing University of Aeronautics and Astronautics), Jie Qin (Nanjing University of Aeronautics and Astronautics)

CodeObject DetectionDomain AdaptationKnowledge DistillationConvolutional Neural NetworkImageVideo

🎯 What it does: Proposes a unified unsupervised salient object detection framework capable of migrating from natural static image (NSI) tasks to non-NSI tasks such as video SOD and remote sensing image SOD.

Unlearning from Weakly Supervised Learning

Yi Tang (Southeast University), Min-Ling Zhang (Southeast University)

CodeSafty and PrivacyImage

🎯 What it does: This paper proposes a machine forgetting method called UDRU for weakly supervised learning, achieving data forgetting by constructing a unified distribution target.

VCformer: Variable Correlation Transformer with Inherent Lagged Correlation for Multivariate Time Series Forecasting

Yingnan Yang (Shenzhen University), Jianyong Chen (Shenzhen University)

CodeTransformerTime Series

🎯 What it does: Propose the VCformer model, combining Variable Correlation Attention (VCA) and Koopman Temporal Detector (KTD) to achieve multivariate time series forecasting.

Vertical Symbolic Regression via Deep Policy Gradient

Nan Jiang (Purdue University), Yexiang Xue (Purdue University)

CodeOptimizationComputational EfficiencyAI Code AssistantRecurrent Neural NetworkReinforcement LearningTabularPhysics RelatedOrdinary Differential Equation

🎯 What it does: Propose a framework that combines vertical symbolic regression with deep policy gradients, capable of incrementally constructing symbolic equations in multivariate scenarios.

VF-Detector: Making Multi-Granularity Code Changes on Vulnerability Fix Detector Robust to Mislabeled Changes

Zhenkan Fu (Dalian Maritime University), He Jiang (Dalian University of Technology)

CodeClassificationRepresentation LearningData-Centric LearningConvolutional Neural NetworkRecurrent Neural NetworkTransformerLarge Language ModelSupervised Fine-TuningTextSequential

🎯 What it does: This paper proposes VF-Detector, which automatically identifies vulnerability fixes in software repair commits using a framework that combines multi-granularity code embeddings with confidence learning for noise removal.

Vision-fused Attack: Advancing Aggressive and Stealthy Adversarial Text against Neural Machine Translation

Yanni Xue (Beihang University), Xianglong Liu (Beihang University)

CodeAdversarial AttackTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelImageTextMultimodality

🎯 What it does: Propose the Vision-fused Attack (VFA) framework, which leverages visual information and semantic space to generate more aggressive and harder-to-be-noticed-by-humans adversarial text for attacking neural machine translation models.

What Hides behind Unfairness? Exploring Dynamics Fairness in Reinforcement Learning

Zhihong Deng (University of Technology Sydney), Chengqi Zhang (University of Technology Sydney)

CodeReinforcement Learning

🎯 What it does: This paper systematically studies the long-term fairness issue in reinforcement learning from a causal inference perspective, proposing a new fairness concept called dynamic fairness, and combining it with traditional direct and indirect effect decomposition to obtain an identifiable formula; subsequently, based on these theories, the model-based reinforcement learning algorithm InsightFair is constructed, which can actively assess and correct unfairness caused by environmental dynamics during the learning process.

WPML3CP: Wasserstein Partial Multi-Label Learning with Dual Label Correlation Perspectives

Ximing Li (Jilin University), Jihong Ouyang (Jilin University)

CodeClassificationImageTextBiomedical Data

🎯 What it does: Propose a dual label correlation perspective partial multi-label learning method WPML CP 3 based on Wasserstein distance, which jointly learns label confidence and prediction model, and improves model performance through label correlation regularization.

Zero-shot Learning for Preclinical Drug Screening

Kun Li (Macquarie University), Wenbin Hu (Macquarie University)

CodeDomain AdaptationDrug DiscoveryGraph Neural NetworkTransformerBiomedical Data

🎯 What it does: Proposed and implemented a zero-shot learning plugin called MSDA to enhance the prediction performance of drug response prediction models for unknown drugs during the preclinical drug screening phase.

ZeroDDI: A Zero-Shot Drug-Drug Interaction Event Prediction Method with Semantic Enhanced Learning and Dual-modal Uniform Alignment

Ziyan Wang (Huazhong Agricultural University), Wen Zhang (Huazhong Agricultural University)

CodeDrug DiscoveryGraph Neural NetworkTransformerLarge Language ModelContrastive LearningTextGraph

🎯 What it does: Propose the ZeroDDI method for predicting unknown drug-drug interaction events (DDIE) without labeled samples in the training set.

Zeta*-SIPP: Improved Time-Optimal Any-Angle Safe-Interval Path Planning

Yiyuan Zou (Delft University of Technology), Clark Borst (Delft University of Technology)

CodeOptimizationBenchmark

🎯 What it does: In dynamic obstacle environments, the Zeta*-SIPP algorithm was proposed and implemented, combining arbitrary-angle forward expansion and field-of-view (FoV) visibility checks to significantly accelerate TO-AA-SIPP.