AAAI 2025 Papers — Page 24
AAAI Conference on Artificial Intelligence · 3028 papers
Regional Expected Improvement for Efficient Trust Region Selection in High-Dimensional Bayesian Optimization
Nobuo Namura (Fujitsu Limited), Sho Takemori (Fujitsu Limited)
OptimizationTabular
🎯 What it does: This paper proposes a Regional Expectation Improvement (REI) and its variant qREI, aimed at better selecting trust regions in high-dimensional Bayesian optimization to enhance global search capabilities.
RegMixMatch: Optimizing Mixup Utilization in Semi-Supervised Learning
Haorong Han (Beijing Jiaotong University), Zhongyang Yu (Communication University of China)
ClassificationOptimizationTransformerContrastive LearningImage
🎯 What it does: Proposes the RegMixMatch framework, improving the use of Mixup in semi-supervised learning by jointly training with high-confidence cleaned samples and mixed samples, and further utilizing low-confidence samples through class-aware Mixup.
REGNav: Room Expert Guided Image-Goal Navigation
Pengna Li (Xi'an Jiaotong University), Sanping Zhou (Xi'an Jiaotong University)
Robotic IntelligenceConvolutional Neural NetworkRecurrent Neural NetworkReinforcement LearningContrastive LearningImage
🎯 What it does: The REGNav model is proposed, utilizing the unsupervised training of Room Expert to help agents determine whether the target and the observed image are in the same room, thereby improving the efficiency of image target navigation.
Regret Analysis of Multi-task Representation Learning for Linear-Quadratic Adaptive Control
Bruce D. Lee (University of Pennsylvania), Nikolai Matni (Columbia University)
Representation LearningReinforcement LearningTime Series
🎯 What it does: An adaptive linear quadratic regulator (LQR) control algorithm for multi-task shared representation learning is proposed, along with finite sample upper bounds in two environments (identifiable and unidentifiable);
Reinforced Multi-teacher Knowledge Distillation for Efficient General Image Forgery Detection and Localization
Zeqin Yu (Sun Yat-sen University), Yuzhen Lin (Shenzhen University)
Anomaly DetectionKnowledge DistillationConvolutional Neural NetworkReinforcement LearningImage
🎯 What it does: A Reinforced Multi-teacher Knowledge Distillation (Re-MTKD) framework is proposed for image forgery detection and localization tasks.
Reinforcement Active Client Selection for Federated Heterogeneous Graph Learning
Jia Wang (Beijing University of Posts and Telecommunications), Guanhua Ye (Beijing University of Posts and Telecommunications)
Federated LearningGraph Neural NetworkReinforcement LearningGraph
🎯 What it does: A client selection method combining reinforcement learning and active learning, called RAF HGL, is proposed for federated heterogeneous graph learning.
Relation Also Knows: Rethinking the Recall and Editing of Factual Associations in Auto-Regressive Transformer Language Models
Xiyu Liu (Institute of Information Engineering, Chinese Academy of Sciences), Weiping Wang (University of Electronic Science and Technology of China)
TransformerLarge Language ModelText
🎯 What it does: A single knowledge editing method based on relational information, RETS, is proposed, which utilizes the MLP sublayer of the autoregressive Transformer to modify weights at the last position of relational tags, significantly reducing the phenomenon of overgeneralization.
Relation-Aware Equivariant Graph Networks for Epitope-Unknown Antibody Design and Specificity Optimization
Lirong Wu (Zhejiang University), Stan Z. Li (Westlake University)
OptimizationDrug DiscoveryGraph Neural NetworkGraphBiomedical Data
🎯 What it does: A relationship-aware equivariant graph network framework RAAD is proposed to generate antibody CDR sequences and structures in a single step under unknown display conditions, optimizing antibody specificity through comparative constraints.
Relation-aware Hierarchical Prompt for Open-vocabulary Scene Graph Generation
Tao Liu (ShanghaiTech University), Xuming He (ShanghaiTech University)
Object DetectionGenerationTransformerLarge Language ModelPrompt EngineeringVision Language ModelImageText
🎯 What it does: A hierarchical prompting framework (RAHP) is proposed to enhance the text representation and visual-text matching of open vocabulary scene graph generation through entity perception and region perception prompts.
Relational Neurosymbolic Markov Models
Lennert De Smet (KU Leuven), Giuseppe Marra (KU Leuven)
GenerationData SynthesisOptimizationSequential
🎯 What it does: A new relational neural symbolic Markov model (NeSy-MMs) is proposed, which integrates deep sequence models with satisfiable relational logic constraints.
Relaxed Class-consensus Consistency for Semi-supervised Semantic Segmentation
Huayu Mai (University of Science and Technology of China), Feng Wu (University of Science and Technology of China)
SegmentationImage
🎯 What it does: This paper proposes a class consistency constraint based on cooperative game theory in semi-supervised semantic segmentation.
Relaxed Rotational Equivariance via G-Biases in Vision
Zhiqiang Wu (East China Normal University), Xian Wei
ClassificationObject DetectionConvolutional Neural NetworkImage
🎯 What it does: This paper proposes a Relaxed Rotational Equivariant Convolution method (RREConv) in the visual domain, which breaks the strict rotational symmetry by incorporating learnable G-Biases into Group Equivariant Convolution (GConv), constructing RRENet (for classification) and RREDet (for detection) networks;
Relieving Universal Label Noise for Unsupervised Visible-Infrared Person Re-Identification by Inferring from Neighbors
Xiao Teng (National University of Defense Technology), Nan Yin (National University of Defense Technology)
RecognitionRetrievalGraph Neural NetworkContrastive LearningImageMultimodality
🎯 What it does: This paper proposes a method to alleviate global label noise in unsupervised visible-infrared person re-identification (USL-VI-ReID) using neighbor information, which mainly includes the Neighbor-guided Universal Label Calibration (N-ULC) and Neighbor-guided Dynamic Weighting (N-DW) modules.
REM: A Scalable Reinforced Multi-Expert Framework for Multiplex Influence Maximization
Huyen Nguyen (Posts and Telecommunications Institute of Technology), Cuong Pham (Posts and Telecommunications Institute of Technology)
OptimizationGraph Neural NetworkReinforcement LearningMixture of ExpertsAuto EncoderGraph
🎯 What it does: A multi-layer network influence maximization framework named REM has been designed and implemented, integrating VAE encoding of seed sets, RL latent space exploration, mixed expert GNN prediction diffusion, and adaptive enhancement through prioritized replay memory;
ReMask-Animate: Refined Character Image Animation Using Mask-Guided Adapters
Xunzhi Xiang (Guangdong Laboratory of Artificial Intelligence and Digital Economy), Fei Richard Yu (Shenzhen University)
GenerationPose EstimationDiffusion modelImageVideo
🎯 What it does: A ReMask-Animate framework is proposed, utilizing hand, face, and foreground/background masks guided adapters (P-Adapter, S-Adapter, C-Adapter) for pose-controlled portrait animation generation.
RemDet: Rethinking Efficient Model Design for UAV Object Detection
Chen Li (Zhejiang Normal University), Xinzhong Zhu (Zhejiang Normal University)
Object DetectionImage
🎯 What it does: A real-time small target detection model for drone images, RemDet, has been designed, reconstructing the network structure to reduce information loss and improve detection accuracy.
ReMoGPT: Part-Level Retrieval-Augmented Motion-Language Models
Qing Yu (LY Corporation), Kent Fujiwara (LY Corporation)
GenerationRetrievalTransformerContrastive LearningTextMultimodalityRetrieval-Augmented Generation
🎯 What it does: A unified motion-language generation model called ReMoGPT is proposed, and retrieval-augmented generation is achieved through fine-grained body part-level text-motion retrieval (PL-TMR);
REPEAT: Improving Uncertainty Estimation in Representation Learning Explainability
Kristoffer K. Wickstrøm (UiT The Arctic University of Norway), Robert Jenssen (University of Copenhagen)
Explainability and InterpretabilityRepresentation LearningConvolutional Neural NetworkTransformerImage
🎯 What it does: A new unsupervised representation learning interpretability method called REPEAT is proposed, which can provide confidence estimates for pixel importance.
RepeatLeakage: Leak Prompts from Repeating as Large Language Model Is a Good Repeater
Yu Peng (Institute of Information Engineering, Chinese Academy of Sciences), Kai Chen (Institute of Information Engineering, Chinese Academy of Sciences)
TransformerLarge Language ModelPrompt EngineeringText
🎯 What it does: It was found that LLMs primarily rely on context rather than model parameters in repetitive tasks. Based on this phenomenon, the RepeatLeakage method was proposed to achieve high-confidence single leakage of system prompts and dialogue context.
RepFace: Refining Closed-Set Noise with Progressive Label Correction for Face Recognition
Jie Zhang (Southwest Jiaotong University), Zhonglin Sun (Queen Mary University of London)
RecognitionImage
🎯 What it does: The RepFace framework is proposed, which processes closed-set noise through steps such as auxiliary sample cleaning, three types of sample classification, label fusion, and label smoothing, thereby enhancing the robustness of facial recognition models.
Replacing Paths with Connection-Biased Attention for Knowledge Graph Completion
Sharmishtha Dutta (Rensselaer Polytechnic Institute), Charu C. Aggarwal (IBM)
Knowledge DistillationGraph Neural NetworkTransformerGraph
🎯 What it does: This paper studies a knowledge graph completion model called CBLiP that does not use path information. It utilizes connection bias attention and entity role embeddings in a subgraph encoder to complete the induction and propagation tasks of knowledge graph completion.
Replicating Electoral Success
Kiran Tomlinson (Cornell University), Jon Kleinberg (Cornell University)
🎯 What it does: A candidate location replicator dynamics model based on the successful replicator strategy is proposed, and it is theoretically proven that when the number of candidates k ≤ 4, candidate locations will converge to the center, while for k ≥ 5, they will not.
Replication-proof Bandit Mechanism Design with Bayesian Agents
Suho Shin (University of Maryland), MohammadTaghi Hajiaghayi (University of Maryland)
OptimizationReinforcement Learning from Human FeedbackReinforcement Learning
🎯 What it does: This study investigates the problem of designing replication protection mechanisms for multi-armed bandits in the context of Bayesian agents, proposing both single-agent and multi-agent replication protection algorithms.
Representation Learning Based Predicate Invention on Knowledge Graphs
Man Zhu (Nanjing University of Posts and Telecommunications), Jingyu Han (Nanjing University of Science and Technology)
Representation LearningGraph Neural NetworkGraph
🎯 What it does: This paper proposes the knowledge graph predicate invention problem based on representation learning (ReLPI) and designs the SEmPI framework to achieve semantic embedding and existence determination of predicates.
Representation Space Augmentation for Effective Self-Supervised Learning on Tabular Data
Moonjung Eo (LG AI Research), Woohyung Lim (LG AI Research)
Representation LearningTransformerContrastive LearningTabular
🎯 What it does: A representation space enhancement method RaTab for self-supervised learning on tabular data is proposed, utilizing truncated singular value decomposition (SVD) for data augmentation at the representation layer;
Representing Sounds as Neural Amplitude Fields: A Benchmark of Coordinate-MLPs and a Fourier Kolmogorov-Arnold Framework
Linfei Li (Tongji University), Ying Shen (Shanghai Jiao Tong University)
GenerationData SynthesisRepresentation LearningBenchmarkAudio
🎯 What it does: This paper first establishes a benchmark for audio implicit representation based on Coordinate-MLP, and then proposes the Fourier-ASR framework, which includes Fourier-KAN and a frequency adaptive learning strategy to achieve high-quality modeling of continuous audio signals.
Reputation-aware Revenue Allocation for Auction-based Federated Learning
Xiaoli Tang (Nanyang Technological University), Han Yu (Nanyang Technological University)
OptimizationFederated LearningImage
🎯 What it does: Designed and implemented a reputation-based dynamic revenue distribution strategy ARAS-AFL, utilizing Lyapunov optimization for revenue distribution and participant attraction in a federated learning auction platform within a multi-market competitive environment.
ResAdapter: Domain Consistent Resolution Adapter for Diffusion Models
Jiaxiang Cheng (ByteDance Inc), Lean Fu (ByteDance Inc)
GenerationData SynthesisDomain AdaptationDiffusion modelImage
🎯 What it does: Designed and trained a lightweight ResAdapter adapter, enabling any diffusion model to achieve resolution interpolation and extrapolation generation while maintaining the original style domain.
Residual Diffusion Deblurring Model for Single Image Defocus Deblurring
Haoxuan Feng (Hong Kong University of Science and Technology), Lei Zhu (Hong Kong University of Science and Technology)
RestorationDiffusion modelImage
🎯 What it does: A single image defocus deblurring framework based on a conditional diffusion model is proposed, which includes a pre-deblurring module and a residual diffusion deblurring module, achieving high-quality full-focus panoramic images in just 2 to 4 steps.
ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidance
Shuwei Shi (University of Tokyo), Yinqiang Zheng (University of Tokyo)
GenerationData SynthesisSuper ResolutionVision Language ModelDiffusion modelImage
🎯 What it does: This paper presents ResMaster, a training-free patch-based diffusion model that achieves seamless expansion from low-resolution reference images to high-resolution images above 4K through structural guidance and fine-grained guidance.
Resolving Multi-Condition Confusion for Finetuning-Free Personalized Image Generation
Qihan Huang (Zhejiang University), Jie Song (Alibaba Group)
GenerationData SynthesisTransformerContrastive LearningImage
🎯 What it does: For personalized image generation with multiple reference images without fine-tuning, a weighted merge scheme is proposed to address the object confusion problem, and based on this, the model is further fine-tuned to improve generation quality.
Resource Constrained Pathfinding with Enhanced Bidirectional A* Search
Saman Ahmadi (RMIT University), Mahdi Jalili (University of Auckland)
OptimizationGraphBenchmark
🎯 What it does: This paper proposes an improved bidirectional A* algorithm RCEBDA* for solving the Resource-Constrained Shortest Path (RCSP) problem, which can efficiently search and output the optimal path in large-scale networks.
Responsibility-aware Strategic Reasoning in Probabilistic Multi-Agent Systems
Chunyan Mu (University of Aberdeen), Nir Oren (University of Aberdeen)
Reinforcement Learning
🎯 What it does: The PATL+R logic is proposed, extending PATL to include causal responsibility and reward metrics, constructing a parameterized stochastic multi-agent system model, and implementing PSPACE-level model checking and Nash equilibrium strategy synthesis for this logic.
Responsive Dynamic Graph Disentanglement for Metro Flow Forecasting
Qiang Gao (Southwestern University of Finance and Economics), Xueqin Chen (Kash Institute of Electronics and Information Industry)
Graph Neural NetworkGraphTime Series
🎯 What it does: A responsive dynamic graph neural network without a fixed graph structure, ReDyNet, is proposed for accurate prediction of subway passenger flow.
RESQUE: Quantifying Estimator to Task and Distribution Shift for Sustainable Model Reusability
Vishwesh Sangarya (North Carolina State University), Jung-Eun Kim (North Carolina State University)
ClassificationDomain AdaptationComputational EfficiencyConvolutional Neural NetworkTransformerImage
🎯 What it does: A prediction quantifier named RESQUE is proposed to estimate the resource cost required for model retraining in the case of distribution shifts or task changes, and this metric can be obtained with a single forward pass before training.
Restabilizing Diffusion Models with Predictive Noise Fusion Strategy for Image Super-Resolution
Luoqian Jiang (South China University of Technology), Jian Chen (South China University of Technology)
RestorationSuper ResolutionTransformerDiffusion modelImage
🎯 What it does: A Predictive Noise Fusion Strategy (PNFS) is proposed, which stabilizes the performance of diffusion models in image super-resolution tasks by predicting pixel-level errors and fusing different noises.
Rethinking Byzantine Robustness in Federated Recommendation from Sparse Aggregation Perspective
Zhongjian Zhang (Beijing University of Posts and Telecommunications), Chuan Shi (Beijing University of Posts and Telecommunications)
Recommendation SystemFederated LearningAdversarial AttackGraph Neural NetworkTabular
🎯 What it does: This paper systematically studies the robustness of Byzantine attacks in the context of sparse aggregation mechanisms in federated recommendation and proposes various attack strategies called Spattack against this sparse aggregation.
Rethinking Cancer Gene Identification Through Graph Anomaly Analysis
Yilong Zang (Hong Kong Polytechnic University), Junhang Wu (University of Queensland)
RecognitionAnomaly DetectionDrug DiscoveryGraph Neural NetworkTransformerGraphBiomedical Data
🎯 What it does: This paper studies the weight heterogeneity of cancer genes in protein-protein interaction (PPI) networks, discovering that it leads to the abnormal feature of spectral energy 'flattening out'. Based on this, a hierarchical perspective graph neural network (HIPGNN) is proposed for cancer gene identification, utilizing dual perspectives of spectral feature encoding and spatial context decoding.
Rethinking High-speed Image Reconstruction Framework with Spike Camera
Kang Chen (Peking University), Zhaofei Yu (Peking University)
RestorationSpiking Neural NetworkPrompt EngineeringContrastive LearningImage
🎯 What it does: Proposes the SpikeCLIP framework, which utilizes CLIP's text-image alignment as weak supervision to reconstruct high-quality images from low-light spike camera data.
Rethinking Masked Data Reconstruction Pretraining for Strong 3D Action Representation Learning
Tao Gong (University of Science and Technology of China), Nenghai Yu (University of Science and Technology of China)
RecognitionRepresentation LearningTransformerAuto EncoderContrastive LearningVideo
🎯 What it does: A self-supervised pre-training framework that combines masked data reconstruction and contrastive learning for 3D action recognition is proposed.
Rethinking Open-Vocabulary Segmentation of Radiance Fields in 3D Space
Hyunjee Lee (Yonsei University), Youngjung Uh (Yonsei University)
SegmentationNeural Radiance FieldGaussian SplattingPoint CloudMesh
🎯 What it does: Achieving open vocabulary semantic segmentation in 3D radiance fields, this paper proposes transforming the task from 2D rendered views to 3D voxel segmentation, directly applying semantic supervision to 3D points, utilizing SAM to obtain scale-invariant language embeddings, and transferring the learned language fields to 3D Gaussian Splatting for real-time rendering.
Rethinking Pseudo-Label Guided Learning for Weakly Supervised Temporal Action Localization from the Perspective of Noise Correction
Quan Zhang (Tsinghua University), Chun Yuan (Tsinghua University)
RecognitionObject DetectionVideo
🎯 What it does: This study addresses the issue of pseudo-label noise in weakly supervised temporal action localization by proposing a framework called NoCo for gradually correcting pseudo-labels.
Rethinking the Starting Point: Collaborative Pre-Training for Federated Downstream Tasks
Yun-Wei Chu (Purdue University), Christopher G. Brinton (University at Buffalo-SUNY)
ClassificationFederated LearningMeta LearningConvolutional Neural NetworkImage
🎯 What it does: This paper proposes CoPreFL, a method for collaborative pre-training using meta-learning in federated learning downstream tasks, aimed at providing a robust initialization model for any heterogeneous FL tasks that may include unseen labels.
Rethinking Transformer-Based Blind-Spot Network for Self-Supervised Image Denoising
Junyi Li (Harbin Institute of Technology), Wangmeng Zuo (Harbin Institute of Technology)
RestorationKnowledge DistillationTransformerImage
🎯 What it does: A Transformer-based Blind Spot Network (TBSN) is proposed for self-supervised image denoising.
Rethinking U-Net: Task-Adaptive Mixture of Skip Connections for Enhanced Medical Image Segmentation
Zichen Luo (Tianjin University), Biao Sun (Tianjin University)
SegmentationConvolutional Neural NetworkMixture of ExpertsImageBiomedical DataMagnetic Resonance Imaging
🎯 What it does: A TA-MoSC module is proposed based on U-Net, treating skip connections as a task allocation problem, and dynamically assigning appropriate multi-scale features for different decoding stages through a sparse Mixture-of-Experts router, thus achieving adaptive skip connections.
RetouchGPT: LLM-based Interactive High-Fidelity Face Retouching via Imperfection Prompting
Wen Xue (South China University of Technology), Hau-San Wong (City University of Hong Kong)
RestorationGenerationTransformerLarge Language ModelReinforcement LearningPrompt EngineeringImage
🎯 What it does: Proposes RetouchGPT, an interactive high-fidelity facial retouching framework based on large language models;
RETQA: A Large-Scale Open-Domain Tabular Question Answering Dataset for Real Estate Sector
Zhensheng Wang (Beijing Normal University), Weijia Jia (Beijing Normal University)
TransformerLarge Language ModelSupervised Fine-TuningTabularFinance Related
🎯 What it does: The RETQA dataset and SLUTQA framework are proposed for open-domain long table question answering in the real estate industry.
RETRACTED: GEONet: Global Enhancement and Optimization Network for Lane Detection
Suyang Xi (Xiamen University), Xiaoxuan Liang (University of Massachusetts Amherst)
Object DetectionAutonomous DrivingConvolutional Neural NetworkTransformerContrastive LearningImage
🎯 What it does: A global enhancement and optimization network named GEONet is proposed for lane detection in complex scenarios, combining a Global Feature Extraction Module (GFEM) and a Top-level Supplement Module (TTSM), and incorporating Whitened Batch Normalization (WBN), Whitened Contrastive Learning (WCL), as well as novel Angle Loss and GRIoU Loss to improve detection accuracy and robustness.
Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning
Jian Lang (University of Electronic Science and Technology of China), Fan Zhou (Kash Institute of Electronics and Information Industry)
GenerationRetrievalTransformerPrompt EngineeringMultimodalityRetrieval-Augmented Generation
🎯 What it does: A retrieval-augmented dynamic prompt tuning framework RAGPT is designed to address the issue of missing modalities;
Retrieval-Augmented Visual Question Answering via Built-in Autoregressive Search Engines
Xinwei Long (Tsinghua University), Bowen Zhou (Tsinghua University)
GenerationRetrievalTransformerLarge Language ModelReinforcement LearningPrompt EngineeringTextMultimodalityRetrieval-Augmented Generation
🎯 What it does: This paper proposes a unified retrieval-generation framework named ReAuSE for knowledge-driven visual question answering tasks, integrating generative retrieval and answer generation within the same large-scale multimodal language model.
REVECA: Adaptive Planning and Trajectory-Based Validation in Cooperative Language Agents Using Information Relevance and Relative Proximity
SeungWon Seo (Kyung Hee University), HyeongYeop Kang (Korea University)
TransformerLarge Language ModelAgentic AITextChain-of-Thought
🎯 What it does: The REVECA framework is proposed in partially observable multi-agent cooperative environments to achieve efficient memory management, planning, and verification.
Revelations: A Decidable Class of POMDPs with Omega-Regular Objectives
Marius Belly (CNRS, LaBRI, Universite de Bordeaux), Pierre Vandenhove (University of Antwerp)
Reinforcement Learning
🎯 What it does: Weak and strong reveal constraints are defined in partially observable Markov decision processes (POMDPs), and it is proven that under these two types of constraints, almost-sure strategies that satisfy ω-regular objectives (including Parity, Buchi, CoBuchi) can be decided, and can be solved by reducing the POMDP to a belief-supported MDP, with an algorithmic complexity of EXPTIME;
Reverse Distribution Based Video Moment Retrieval for Effective Bias Elimination
Lingdu Kong (Northeastern University), Xiangmin Zhou (RMIT University)
RetrievalGaussian SplattingVideoText
🎯 What it does: A video moment retrieval method based on reverse distribution, ReDis-VMR, and a dynamically scalable adjustment DEA is proposed to eliminate data and model bias and improve retrieval performance.
Reverse Region-to-Entity Annotation for Pixel-Level Visual Entity Linking
Zhengfei Xu (Beijing Institute of Technology), Xin Xin (Tencent)
Object DetectionSegmentationTransformerLarge Language ModelSupervised Fine-TuningImage
🎯 What it does: This paper studies the pixel-level visual entity linking (PL-VEL) task and constructs the MaskOVEN-Wiki large-scale pixel-level annotated dataset.
Revisiting Attention for Multivariate Time Series Forecasting
Haixiang Wu (Jiangsu University)
TransformerTime Series
🎯 What it does: Two novel attention mechanisms, FSatten and SOatten, are proposed and implemented to enhance the performance of multivariate time series forecasting.
Revisiting CAD Model Generation by Learning Raster Sketch
Pu Li (Institute of Automation, Chinese Academy of Sciences), Dong-Ming Yan (Institute of Automation, Chinese Academy of Sciences)
GenerationData SynthesisTransformerDiffusion modelAuto EncoderImage
🎯 What it does: This paper proposes the RECAD framework, which utilizes a representation of rasterized sketches paired with extrusion boxes to generate CAD 3D models through a two-stage diffusion model.
Revisiting Change Captioning from Self-supervised Global-Part Alignment
Feixiao Lv (Institute of Information Engineering), Lihua Jing (Institute of Information Engineering)
TransformerContrastive LearningImage
🎯 What it does: A self-supervised global-partial alignment network (SSGPA) is proposed for image change description.
Revisiting Graph Contrastive Learning on Anomaly Detection: A Structural Imbalance Perspective
Yiming Xu (Xi'an Jiaotong University), Chen Chen (University of Virginia)
Anomaly DetectionGraph Neural NetworkContrastive LearningGraph
🎯 What it does: This paper proposes an anomaly detection framework based on graph contrastive learning (AD-GCL), specifically addressing the poor performance of tail anomaly detection caused by structural imbalance (between head nodes and tail nodes).
Revisiting Interpolation for Noisy Label Correction
Yuanzhuo Xu (Wuhan University), Steve Drew (University of Calgary)
ClassificationData-Centric LearningConvolutional Neural NetworkContrastive LearningImage
🎯 What it does: A label correction method based on gradient scaling theory, DULC, is designed to dynamically determine the interpolation weight of each sample using normalized Jensen-Shannon divergence, and to enhance the model's memory of true labels through weak/strong data augmentation and prediction sharpening.
Revisiting Multimodal Emotion Recognition in Conversation from the Perspective of Graph Spectrum
Wei Ai (Central South University of Forestry and Technology), Keqin Li (State University of New York)
RecognitionRecurrent Neural NetworkGraph Neural NetworkContrastive LearningMultimodality
🎯 What it does: Proposes the GS-MCC framework, which constructs a multimodal interaction graph through a sliding window, uses Fourier graph operators to extract high and low-frequency semantics, and employs contrastive learning to achieve semantic consistency and complementarity, ultimately performing sentiment prediction in a single MLP.
Revisiting Multimodal Fusion for 3D Anomaly Detection from an Architectural Perspective
Kaifang Long (Northeastern University), Zhichao Lu (City University of Hong Kong)
Anomaly DetectionNeural Architecture SearchMultimodalityPoint Cloud
🎯 What it does: This paper systematically analyzes the impact of multimodal fusion architecture on 3D anomaly detection performance, proposing a two-layer fusion search space and implementing 3D-ADNAS.
Revisiting Projection-Free Online Learning with Time-Varying Constraints
Yibo Wang (Nanjing University), Lijun Zhang (Nanjing University)
OptimizationReinforcement Learning
🎯 What it does: This paper studies constrained online convex optimization problems and proposes a new projection-free method to handle time-varying constraints.
Revisiting Tampered Scene Text Detection in the Era of Generative AI
Chenfan Qu (South China University of Technology), Lianwen Jin (South China University of Technology)
Object DetectionSegmentationTransformerImageBenchmark
🎯 What it does: This paper proposes an open-set text detection task for forged scenes, constructs a high-quality benchmark dataset covering eight generative text editing models, and introduces a texture jitter pre-training and difference-aware forensics framework to enhance the model's detection capability for unknown forgeries.
Revolutionizing Encrypted Traffic Classification with MH-Net: A Multi-View Heterogeneous Graph Model
Haozhen Zhang (Tsinghua University), Ye Zhang (Nanjing University of Posts and Telecommunications)
ClassificationSafty and PrivacyGraph Neural NetworkContrastive LearningGraph
🎯 What it does: Proposes MH-Net, which utilizes multi-view heterogeneous flow graphs for packet-level and flow-level classification of encrypted traffic.
ReX: A Framework for Incorporating Temporal Information in Model-Agnostic Local Explanation Techniques
Junhao Liu (Peking University), Xin Zhang (Peking University)
Anomaly DetectionExplainability and InterpretabilityRecurrent Neural NetworkTransformerTextTime SeriesSequentialElectrocardiogram
🎯 What it does: For variable-length sequence models (such as RNNs and Transformers), the REX framework is proposed to embed temporal information into local model-agnostic interpretability techniques (Anchors, LIME, Kernel SHAP).
RFL: Simplifying Chemical Structure Recognition with Ring-Free Language
Qikai Chang (University of Science and Technology of China), Jinshui Hu (iFLYTEK Research)
RecognitionRecurrent Neural NetworkImage
🎯 What it does: This paper proposes a Ring-Free Language (RFL) that decouples chemical structure diagrams into molecular skeletons, individual ring, and branch information, and based on this, designs a unified Molecular Skeleton Decoder (MSD) for end-to-end recognition.
RGBT Tracking via All-layer Multimodal Interactions with Progressive Fusion Mamba
Andong Lu (Anhui University), Bin Luo (Anhui University)
Object TrackingTransformerMultimodality
🎯 What it does: This paper proposes a full-layer multimodal interaction network called AINet, which utilizes Mamba to achieve differential fusion and dynamic scanning, significantly enhancing RGB-TIR target tracking performance.
RHanDS: Refining Malformed Hands for Generated Images with Decoupled Structure and Style Guidance
Chengrui Wang (Taobao and Tmall Group of Alibaba), Bo Zheng (Taobao and Tmall Group of Alibaba)
GenerationData SynthesisPose EstimationDiffusion modelImageMesh
🎯 What it does: A RHanDS framework based on diffusion models is proposed to correct deformed hand structures in generated images while maintaining the original style.
RhythmMamba: Fast, Lightweight, and Accurate Remote Physiological Measurement
Bochao Zou (University of Science and Technology Beijing), Huimin Ma (China Academy of Electronics and Information Technology)
Computational EfficiencyConvolutional Neural NetworkTransformerVideoTime Series
🎯 What it does: Designed and implemented RhythmMamba, an efficient remote photoplethysmography (rPPG) signal extraction framework based on state space models.
RI-MAE: Rotation-Invariant Masked AutoEncoders for Self-Supervised Point Cloud Representation Learning
Kunming Su (South China University of Technology), Kun Hu (University of Sydney)
Representation LearningTransformerAuto EncoderPoint Cloud
🎯 What it does: Designed a rotation-invariant masked autoencoder (RI-MAE) for unsupervised point cloud representation learning.
Riemann-based Multi-scale Attention Reasoning Network for Text-3D Retrieval
Wenrui Li (Harbin Institute of Technology), Xiaopeng Fan (Peng Cheng Laboratory)
RetrievalContrastive LearningTextPoint Cloud
🎯 What it does: The RMARN model is proposed, utilizing Riemannian geometry and multi-scale attention to achieve retrieval alignment between text and 3D point clouds.
Riemannian Geometric-based Meta Learning
JuneYoung Park (Opt-AI Inc.), Jang-Hwan Choi (Ewha Womans University)
Meta LearningImage
🎯 What it does: A Stiefel-MAML meta-learning framework based on Riemannian geometry is proposed, which achieves orthogonality constraints on parameters through gradient descent on the Stiefel manifold, enhancing the adaptation speed and accuracy of few-shot learning.
RILQ: Rank-Insensitive LoRA-Based Quantization Error Compensation for Boosting 2-Bit Large Language Model Accuracy
Geonho Lee (Hanyang University), Jungwook Choi (Sogang University)
TransformerLarge Language ModelText
🎯 What it does: To address the accuracy decline of LLMs caused by 2-bit weight quantization, this paper proposes the Rank-Insensitive LoRA-based Quantization Error Compensation (RILQ) method, which utilizes model-level activation gap loss to achieve collaborative compensation for low-rank adapters.
RingFormer: A Ring-Enhanced Graph Transformer for Organic Solar Cell Property Prediction
Zhihao Ding (Hong Kong Polytechnic University), Chen Jason Zhang (Hong Kong Polytechnic University)
Graph Neural NetworkTransformerGraph
🎯 What it does: Developed RingFormer, a graph Transformer framework that combines atomic layers, ring layers, and interlayers for predicting the properties of organic solar cell (OSC) molecules.
Risk-averse Total-reward MDPs with ERM and EVaR
Xihong Su (University of New Hampshire), Julien Grand-Clément (HEC Paris)
OptimizationReinforcement LearningTabularFinance Related
🎯 What it does: This paper studies the introduction of Entropy Risk Measure (ERM) and Entropy Value at Risk (EVaR) as risk measures within the framework of Total Return Criterion (TRC) in finite Markov Decision Processes (MDPs), and proves the existence of an optimal deterministic stationary policy in transient MDPs.
RLPF: Reinforcement Learning from Prediction Feedback for User Summarization with LLMs
Jiaxing Wu (Google DeepMind), Jun Xie (Google DeepMind)
Recommendation SystemReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningText
🎯 What it does: A reinforcement learning framework based on predictive feedback, RLPF, is proposed for generating concise and readable user summaries that can enhance downstream task performance;
RMath: A Logic Reasoning-Focused Datasets Toward Mathematical Multistep Reasoning Tasks
Ziyi Hu (National University of Defense Technology), Yiping Song (National University of Defense Technology)
TransformerLarge Language ModelPrompt EngineeringText
🎯 What it does: A mathematical reasoning dataset RMath for multi-step logical reasoning has been constructed, and a standardized reasoning framework and annotation scheme based on propositional logic have been proposed for training and evaluating the logical reasoning capabilities of LLMs.
Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation
Miran Heo (Yonsei University), Joon-Young Lee (Adobe Research)
Object DetectionSegmentationTransformerVideo
🎯 What it does: This paper proposes the RoCoVIS online video instance segmentation model, which improves pixel-level and instance-level consistency through instance mask propagation.
Robust Graph Based Social Recommendation Through Contrastive Multi-View Learning
Fei Xiong (Beijing Jiaotong University), Liang Wang (Northwestern Polytechnical University)
Recommendation SystemGraph Neural NetworkContrastive LearningGraph
🎯 What it does: The RGCML framework is proposed, which combines social relationship denoising, global intent modeling, personalized information fusion, and global-local contrastive learning to enhance the quality of social recommendations.
Robust Heterogeneous Graph Classification for Molecular Property Prediction with Information Bottleneck
Zhibin Ni (Tsinghua University), Xibin Zhao (Tsinghua University)
ClassificationAdversarial AttackDrug DiscoveryGraph Neural NetworkGraph
🎯 What it does: This paper proposes a Robust Heterogeneous Graph Classification framework (RHGC) for molecular property prediction.
Robust Image Hashing Based on Contrastive Masked Autoencoder with Weak-Strong Augmentation Alignment
Cundian Yang (Peking University), Xiyao Liu (Central South University)
RecognitionRetrievalAnomaly DetectionTransformerAuto EncoderContrastive LearningImage
🎯 What it does: A robust image hashing framework called CMAA based on end-to-end contrastive masked autoencoders is proposed, which enhances robustness against large-scale and mixed attacks by combining weak-strong augmentation alignment, ViT masks, and non-parametric quantization.
Robust Logit Adjustment for Learning with Long-Tailed Noisy Data
MingCai Chen (Nanjing University of Posts and Telecommunications), Chongjun Wang (Nanjing University)
ClassificationData-Centric LearningImage
🎯 What it does: Proposes the Robust Logit Adjustment method, which first performs label refurbishment on noisy labels, and then uses online updated label frequencies for Logit adjustment to address the data learning problem of coexisting long tails and noise.
Robust Performance Incentivizing Algorithms for Multi-Armed Bandits with Strategic Agents
Seyed A. Esmaeili (University of Chicago), Aleksandrs Slivkins (Microsoft Research)
OptimizationReinforcement Learning
🎯 What it does: This paper proposes a class of performance-incentivizing and robust multi-armed bandit algorithms for dealing with worker agents exhibiting strategic behavior.
Robust SAM: On the Adversarial Robustness of Vision Foundation Models
Jiahuan Long (Shanghai Jiao Tong University), Xiaoqian Chen (Chinese Academy of Military Science)
SegmentationAdversarial AttackConvolutional Neural NetworkSupervised Fine-TuningImageVideo
🎯 What it does: This paper proposes Cross-Prompt Adversarial Attack (CPA) and SVD-based Low-Parameter Defense (RobustSAM) to evaluate and enhance the robustness of image and video segmentation models SAM and SAM 2.
Robust Self-Paced Hashing for Cross-Modal Retrieval with Noisy Labels
Ruitao Pu (Sichuan University), Dezhong Peng (Sichuan National Innovation New Vision UHD Video Technology Co., Ltd.)
RetrievalContrastive LearningMultimodality
🎯 What it does: A robust self-paced hashing framework RSHNL is proposed to handle noisy labeled data in cross-modal retrieval.
Robust Tracking via Mamba-based Context-aware Token Learning
Jinxia Xie (Guangxi Normal University), Shuxiang Song (Wuzhou University)
Object TrackingVideo
🎯 What it does: We propose TemTrack, a visual tracker that separates temporal information learning from appearance modeling, capturing target appearance changes and motion trends through trajectory tokens in a sliding window and a mamba-cross attention module.
RoDA: Robust Domain Alignment for Cross-Domain Retrieval Against Label Noise
Ziniu Yin (Sichuan University), Xu Wang (Sichuan University)
RetrievalDomain AdaptationConvolutional Neural NetworkImage
🎯 What it does: This study proposes the RoDA framework to address the issue of label noise in cross-domain image retrieval.
RoPaSS: Robust Watermarking for Partial Screen-Shooting Scenarios
Zehua Ma (Anhui Province Key Laboratory of Digital Security, University of Science and Technology of China), Weiming Zhang (Anhui Province Key Laboratory of Digital Security, University of Science and Technology of China)
RestorationConvolutional Neural NetworkAuto EncoderImage
🎯 What it does: Proposes RoPaSS, a robust watermarking scheme capable of synchronously extracting watermarks in partial screen capture scenarios.
RouterRetriever: Routing over a Mixture of Expert Embedding Models
Hyunji Lee (KAIST AI), Kyle Lo (Allen Institute for AI)
RetrievalMixture of ExpertsTextBenchmark
🎯 What it does: Proposes the RouterRetriever model, which utilizes domain experts and embedding similarity routing for retrieval;
RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data
Chenglong Wang (Northeastern University), Jingbo Zhu (Northeastern University)
Recommendation SystemReinforcement LearningVision Language ModelImageText
🎯 What it does: Through a three-stage progressive training and optimal transport preference data selection, the visual reward model is enhanced using text preference data, allowing large visual language models to better align with human preferences.
RP-PGD: Boosting Segmentation Robustness with a Region-and-Prototype Based Adversarial Attack
Yuxuan Zhang (University of Science and Technology of China), Yinxing Xue (Guangzhou University)
SegmentationAdversarial AttackImage
🎯 What it does: This paper studies an adversarial attack method for semantic segmentation called RP-PGD, which generates stronger adversarial samples using region partitioning and prototype perturbation, and employs them for adversarial training to enhance model robustness.
RRT-MVS: Recurrent Regularization Transformer for Multi-View Stereo
Jianfei Jiang (University of Science and Technology Beijing), Huimin Ma (University of Science and Technology Beijing)
Depth EstimationTransformerPoint Cloud
🎯 What it does: A Recurrent Regularization Transformer (RRT) for multi-view stereo reconstruction has been designed and implemented, generating more accurate depth maps by globally regularizing the cost volume in both depth and spatial dimensions.
Rule-Guided Graph Neural Networks for Explainable Knowledge Graph Reasoning
Zhe Wang (Griffith University), Zhiqiang Zhuang (Tianjin University)
Explainability and InterpretabilityGraph Neural NetworkGraphBenchmark
🎯 What it does: This paper proposes a rule-guided graph neural network for interpretable knowledge graph reasoning, which can utilize existing rules to guide learning during the training process, and directly extract high-confidence rules from the network parameters after the model training is completed.
RUNA: Object-Level Out-of-Distribution Detection via Regional Uncertainty Alignment of Multimodal Representations
Bin Zhang (Wuhan National Laboratory for Optoelectronics), Jianzong Wang (Wuhan National Laboratory for Optoelectronics)
Object DetectionDomain AdaptationAnomaly DetectionTransformerSupervised Fine-TuningVision Language ModelContrastive LearningImageMultimodality
🎯 What it does: This paper proposes an object-level OOD detection framework called RUNA, which is based on dual encoders and region uncertainty alignment, and enhances detection performance through few-shot fine-tuning.
Runtime Analysis for Multi-Objective Evolutionary Algorithms in Unbounded Integer Spaces
Benjamin Doerr (Institut Polytechnique de Paris), Günter Rudolph (Dortmund University)
Optimization
🎯 What it does: The paper provides a theoretical analysis of the running time of multi-objective evolutionary algorithms (SEMO and GSEMO) in unbounded integer search spaces, considering three different mutation strengths (simple, exponential tail distribution, and power-law distribution), and evaluates their performance on a new bi-objective benchmark function.
S-INF: Towards Realistic Indoor Scene Synthesis via Scene Implicit Neural Field
Zixi Liang (University of Electronic Science and Technology of China), Lixin Duan (University of Electronic Science and Technology of China)
GenerationData SynthesisTransformerNeural Radiance FieldPoint CloudMesh
🎯 What it does: A method for indoor scene synthesis called S-INF (Scene Implicit Neural Field) is proposed, which utilizes implicit neural fields to simultaneously learn the relationships of scene layout and detailed objects, generating more realistic and stylistically consistent 3D indoor scenes.
S^3cMath: Spontaneous Step-Level Self-Correction Makes Large Language Models Better Mathematical Reasoners
Yuchen Yan (Zhejiang University), Jian Shao (Peking University)
TransformerLarge Language ModelSupervised Fine-TuningText
🎯 What it does: Proposed the spontaneous one-step self-correction ability of LLM, enabling the model to identify and correct errors in real-time during the reasoning process.
S²DN: Learning to Denoise Unconvincing Knowledge for Inductive Knowledge Graph Completion
Tengfei Ma (Hunan University), Xiangxiang Zeng (Hunan University)
Recommendation SystemKnowledge DistillationGraph Neural NetworkGraph
🎯 What it does: A Semantic Structure-aware Denoising Network (S²DN) is proposed for Inductive Knowledge Graph Completion (Inductive KGC), which enhances inference on new entities by smoothing semantic similarity relationships in subgraphs and adaptively filtering structures.
S²MILE: Semantic-and-Structure-Aware Music-Driven Lyric Generation
Mu You (University of Science and Technology of China), Linli Xu (University of Science and Technology of China)
GenerationTransformerLarge Language ModelSupervised Fine-TuningTextAudio
🎯 What it does: This paper proposes an end-to-end music-driven lyric generation model S²MILE, which can automatically generate lyrics that are structurally and semantically consistent from complete multi-track multi-voice music.
S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging
Yimu Pan (Pennsylvania State University), James Z. Wang (Pennsylvania State University)
SegmentationDomain AdaptationDiffusion modelMultimodalityBiomedical DataMagnetic Resonance ImagingComputed Tomography
🎯 What it does: A semantic stacking (S2S2) training strategy is proposed as an add-on to existing segmentation models to enhance the robustness of the models.