Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks211 from public data
- Multi-scale Residual Network for Image Super-Resolution925
A novel multi-scale residual network (MSRN) to fully exploit the image features, which outperform most of the state-of-the-art methods.
- Transformer for Single Image Super-Resolution677
This paper proposes a novel Efficient Super-Resolution Transformer (ESRT) for SISR, a hybrid model, which consists of a Lightweight CNN Backbone and a Light Weight Transformer Backbone that achieves competitive results with low computational cost.
- Masked Autoencoders that Listen524
The Audio-MAE, a simple extension of image-based Masked Autoencoders to self-supervised representation learning from audio spectrograms, sets new state-of-the-art performance on six audio and speech classification tasks, outperforming other recent models that use external supervised pre-training.
- Very deep convolutional neural networks for raw waveforms399
This work proposes very deep convolutional neural networks that directly use time-domain waveforms as inputs that are efficient to optimize over very long sequences, necessary for processing acoustic waveforms.
- Learning Joint Embedding with Multimodal Cues for Cross-Modal Video-Text Retrieval288
This paper proposes a novel framework that simultaneously utilizes multi-modal features (different visual characteristics, audio inputs, and text) by a fusion strategy for efficient retrieval and explores several loss functions in training the embedding.
- AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea261
AnyEdit is presented, a comprehensive multi-modal instruction editing dataset, comprising 2.5 million high-quality editing pairs spanning over 20 editing types and five domains, and shows that AnyEdit consistently boosts the performance of diffusion-based editing models.
- 247
- A Comparison of Five Multiple Instance Learning Pooling Functions for Sound Event Detection with Weak Labeling203
This paper builds a neural network called TALNet, which is the first system to reach state-of-the-art audio tagging performance on Audio Set, while exhibiting strong localization performance on the DCASE 2017 challenge at the same time.
- Lightweight Real-Time Semantic Segmentation Network With Efficient Transformer and CNN184
This work proposes a lightweight real-time semantic segmentation network called LETNet that combines a U-shaped CNN with Transformer effectively in a capsule embedding style to compensate for respective deficiencies.
- Universal Phone Recognition with a Multilingual Allophone System183
This work proposes a joint model of both language-independent phone and language-dependent phoneme distributions that can build a (nearly-)universal phone recognizer that, when combined with the PHOIBLE [1] large, manually curated database of phone inventories, can be customized into 2,000 language dependent recognizers.
- 181
- Structure-Preserving Deraining with Residue Channel Prior Guidance180
A Structure-Preserving Deraining Network (SPDNet) with RCP guidance is proposed, which directly generates high-quality rain-free images with clear and accurate structures under the guidance of RCP but does not rely on any rain-generating assumptions.
- CTCNet: A CNN-Transformer Cooperation Network for Face Image Super-Resolution164
This work proposes an efficient CNN-Transformer Cooperation Network (CTCNet) for face super-resolution tasks, which uses the multi-scale connected encoder-decoder architecture as the backbone and devise a novel Local-Global Feature Cooperation Module (LGCM) to promote the consistency of local facial detail and global facial structure restoration simultaneously.
- 164
- Momentor: Advancing Video Large Language Model with Fine-Grained Temporal Reasoning156
The proposed Momentor, a Video-LLM capable of accomplishing fine-grained temporal understanding tasks, is designed and trained on Moment-10M, a large-scale video instruction dataset with segment-level instruction data, enabling it to perform segment-level reasoning and localization.
- FBSNet: A Fast Bilateral Symmetrical Network for Real-Time Semantic Segmentation154
This article proposes a Fast Bilateral Symmetrical Network (FBSNet), a symmetrical encoder-decoder structure with two branches, semantic information branch and spatial detail branch that can strike a good balance between accuracy and efficiency.
- A comparison of Deep Learning methods for environmental sound detection149
This work presents a comparison of several state-of-the-art Deep Learning models on the IEEE challenge on Detection and Classification of Acoustic Scenes and Events (DCASE) 2016 challenge task and data, classifying sounds into one of fifteen common indoor and outdoor acoustic scenes.
- 146
- 146
- 139
- Progressive Feature Fusion Network for Realistic Image Dehazing133
An U-Net like encoder-decoder deep network via progressive feature fusions has been proposed to directly learn highly nonlinear transformation function from observed hazy image to haze-free ground-truth and can achieve superior performance when compared with popular state-of-the-art methods.
- MDCN: Multi-Scale Dense Cross Network for Image Super-Resolution126
A Multi-scale Dense Cross Network (MDCN), which achieves great performance with fewer parameters and less execution time and achieves competitive results in SISR, especially in the reconstruction task with multiple upsampling factors.
- Feature Distillation Interaction Weighting Network for Lightweight Image Super-resolution122
This work proposes a lightweight yet efficient Feature Distillation Interaction Weighted Network (FDIWN), which utilizes a series of specially designed Feature Shuffle Weighted Groups as the backbone, and several novel mutual Wide-residual Distillation interaction Blocks form an FSWG.
- Fine-Grained Semantically Aligned Vision-Language Pre-Training111
This paper introduces LOUPE, a fine-grained semantically aLigned visiOn-langUage PrE-training framework, which learns fine- grained semantic alignment from the novel perspective of game-theoretic interactions, and proposes an uncertainty-aware neural Shapley interaction learning module.
- Compositional Temporal Grounding with Structured Variational Cross-Graph Correspondence Learning105
A variational cross-graph reasoning framework that explicitly decomposes video and language into multiple structured hierarchies and learns fine-grained semantic correspondence among them is proposed and empirically finds that they fail to generalize to queries with novel combinations of seen words.
- Recent progress of preparation of branched poly(lactic acid) and its application in the modification of polylactic acid materials90
Branched PLA has excellent compatibility with PLA, which has important research value in regulating the structure and properties of PLA materials.
- 90
- Cross-Receptive Focused Inference Network for Lightweight Image Super-Resolution89
A novel Cross-receptive Field Guided Transformer is designed to enable the selection of contextual information required for reconstruction by using a modulated convolutional kernel that understands the current semantic information and exploits the information interaction within different self-attention.
- 87
- 86
- WorldGPT: Empowering LLM as Multimodal World Model80
WorldGPT is introduced, a generalist world model built upon Multimodal Large Language Model (MLLM) that acquires an understanding of world dynamics through analyzing millions of videos across various domains and is integrated with a novel cognitive architecture that combines memory offloading, knowledge retrieval, and context reflection.
- Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation80
This paper proposes a novel unsupervised reinforcement learning approach to learn transferable meta-skills from unannotated environments without any supervisory signals, and shows that this method learns transferable motor primitives for visual navigation.
- A Systematic Survey of Deep Learning-Based Single-Image Super-Resolution79
This survey gives an overview of DL-based SISR methods and group them according to their design targets and presents the reconstruction results of some classic SISR methods to intuitively know their performance.
- 70
- Feedback Network for Mutually Boosted Stereo Image Super-Resolution and Disparity Estimation70
A Stereo Super-Resolution and Disparity Estimation Feedback Network (SSRDE-FNet), which simultaneously handles the stereo image super-resolution and disparity estimation in a unified framework and interact them with each other to further improve their performance.
- Visually-Prompted Language Model for Fine-Grained Scene Graph Generation in an Open World63
A Cross-modal prediCate boosting (CaCao) framework, where a visually-prompted language model is learned to generate diverse fine-grained predicates in a low-resource way for open-world predicate scene graph generation (Epic).
- Efficient and Accurate Multi-Scale Topological Network for Single Image Dehazing60
A multi-scale topological network that provides a large number of search paths that enable the network to extract abundant image features as well as strong fault tolerance and robustness and adaptively select important features and ignore interference information when fusing different scale representations is proposed.
- 58
- InstructVid2Vid: Controllable Video Editing with Natural Language Instructions56
Experimental results demonstrate that InstructVid2Vid is capable of generating high-quality, temporally coherent videos and performing diverse edits, including attribute editing, background changes, and style transfer, and underscores the versatility and effectiveness of the proposed method.
- MAGIC: Multimodal relAtional Graph adversarIal inferenCe for Diverse and Unpaired Text-Based Image Captioning52
This work proposes the Multimodal relAtional Graph adversarIal InferenCe (MAGIC) framework, a framework that can adaptively construct multiple multimodal relational graphs of images and model complex relationships among graphs to represent descriptive diversity and validate the effectiveness of MAGIC in generating diverse captions from different relational information items of an image.
- 49
- 48
- Efficient Image Super-Resolution With Feature Interaction Weighted Hybrid Network46
This work proposes a Feature Interaction Weighted Hybrid Network (FIWHN), which comprises a series of Wide-residual Distillation Interaction Block (WDIB) as the backbone, and incorporates a Transformer and explores a novel architecture to combine CNN and Transformer.
- Revisiting the Domain Shift and Sample Uncertainty in Multi-source Active Domain Transfer43
A Dynamic integrated uncertainty valuation framework (Detective) that comprehensively consider the domain shift between multi-source domains and target domain to detect the informative target samples and mitigating uncertainty miscalibration is proposed.
- Auto-Encoding Morph-Tokens for Multimodal LLM41
This work proposes encoding images into morph-tokens to serve a dual purpose: for comprehension, they act as visual prompts instructing MLLM to generate texts; for generation, they take on a different, non-conflicting role as complete visual-tokens for image reconstruction, where the missing visual cues are recovered by the MLLM.
- Gradient-Regulated Meta-Prompt Learning for Generalizable Vision-Language Models41
A novel Gradient-RegulAted Meta-prompt learning (GRAM) framework that jointly meta-learns an efficient soft prompt initialization for better adaptation and a lightweight gradient regulating function for strong cross-domain generalizability in a meta-learning paradigm using only the unlabeled image-text pre-training data.
- 39
- 39
- 38
- Weakly Supervised Lesion Detection and Diagnosis for Breast Cancers With Partially Annotated Ultrasound Images37
A novel Two-Stage Detection and Diagnosis Network (TSDDNet) is proposed based on weakly supervised learning to improve diagnostic accuracy of the ultrasound-based CAD for breast cancers, suggesting promising application potential.
- Review of Quaternion-Based Color Image Processing Methods37
Past methods are reviewed and categorized to indicate that the quaternion representation method is effective, and models and methods based on it have rapidly developed.
- 36
- Towards Zero-Shot Learning for Automatic Phonemic Transcription36
This model is able to recognize unseen phonemes in the target language without any training data and achieves 7.7% better phoneme error rate on average over a standard multilingual model.
- NTIRE 2022 Challenge on Stereo Image Super-Resolution: Methods and Results35
The 1st NTIRE challenge on stereo image super-resolution (restoration of rich details in a pair of low-resolution stereo images) with a focus on new solutions and results establishes a new benchmark for stereo image SR.
- STEP: Enhancing Video-LLMs’ Compositional Reasoning by Spatio-Temporal Graph-guided Self-Training33
STEP, a novel graph-guided self-training method that enables Video-LLMs to generate reasoning-rich fine-tuning data from any raw videos to improve itself, achieves superior performance with a minimal amount of self-generated rationale-enriched training samples in both compositional reasoning and comprehensive understanding benchmarks, highlighting the broad applicability and vast potential.
- HumanEdit: A High-Quality Human-Rewarded Dataset for Instruction-based Image Editing32
High-quality, human-rewarded dataset specifically designed for instruction-guided image editing, enabling precise and diverse image manipulations through open-form language instructions, set a new versatile benchmark for instructional image editing datasets.
- Pseudo-Data Based Self-Supervised Federated Learning for Classification of Histopathological Images32
A novel Barlow Twins based FL (FL-BT) algorithm is proposed to improve the local training for the CAD models in each center by conducting model contrastive learning, which benefits the optimization of the global model in the FL procedure.
- Joint embeddings with multimodal cues for video-text retrieval31
This paper proposes a framework that simultaneously utilizes multimodal visual cues by a “mixture of experts” approach for retrieval and conducts extensive experiments to verify that the system is able to boost the performance of the retrieval task compared to the state of the art.
- EyecareGPT: Boosting Comprehensive Ophthalmology Understanding with Tailored Dataset, Benchmark and Model30
The Eyecare Kit is proposed, which systematically tackles three key challenges of intelligent ophthalmic diagnosis with the tailored dataset, benchmark and model, and develops the EyecareGPT, optimized for fine-grained ophthalmic visual understanding thoroughly.
- 30
- A fine-grained orthodontics segmentation model for 3D intraoral scan data29
This work focuses on deformed teeth segmentation and provides a fine-grained tooth segmentation dataset (3D-IOSSeg), and proposes a fast graph convolutional network for 3D tooth segmentation named Fast-TGCN, which can quickly and accurately segment teeth from the mouth with complex structures and outperforms other methods in various evaluation metrics.
- 29
- HighEr-Resolution Network for Image Demosaicing and Enhancing29
A HighEr-Resolution Network (HERN) to fully learning global information in high-resolution image patches by combining global-aware features and multi-scale features, which achieves state-of-the-art performance on the AIM2019 RAW to RGB mapping challenge.
- Cross Paradigm Representation and Alignment Transformer for Image Deraining28
This work proposes a novel Cross Paradigm Representation and Alignment Transformer, which aligns and interacts with features in a two-stage progressive manner, enabling adaptive guidance and complementarity, and introduces the Adaptive Alignment Frequency Module (AAFM), which reduces the information gap within and between paradigms.
- 28
- Variational Cross-Graph Reasoning and Adaptive Structured Semantics Learning for Compositional Temporal Grounding28
A variational cross-graph reasoning framework is proposed that explicitly decomposes video and language into hierarchical semantic graphs, respectively, and learns fine-grained semantic correspondence between the two graphs, and introduces a novel adaptive structured semantics learning approach to derive the structure-informed and domain-generalizable graph representations.
- Adaptive Hierarchical Graph Reasoning with Semantic Coherence for Video-and-Language Inference28
An adaptive hierarchical graph network is proposed that achieves in-depth understanding of the video over complex interactions andsemantic coherence learning is introduced to explicitly encourage the semantic coherence of the adaptive hierarchicalgraph network from three hierarchies.
- 27
- Survey on Deep Face Restoration: From Non-blind to Blind and Beyond26
The prevalent factors responsible for real-world LQ images and degradation techniques used to synthesize LQ images are examined and degradation techniques used to synthesize LQ images are introduced.
- Evaluating the Robustness of Multimodal Agents Against Active Environmental Injection Attacks26
This work proposes AEIA-MN, an attack scheme that exploits interaction vulnerabilities in mobile operating systems to assess the robustness of MLLM-based agents, and identifies two critical security vulnerabilities in the agent's task execution process, which increase susceptibility to AEIA attacks during reasoning.
- Multi-Scale Efficient Graph-Transformer for Whole Slide Image Classification26
This work designs an EGT to efficiently learn the local-global information of patch tokens, which integrates the graph representation into Transformer to capture spatial-related information of WSIs and proposes a novel MFFM to alleviate the semantic gap among different resolution patches during feature fusion.
- 24
- Enhancing feature discrimination with pseudo-labels for foundation model in segmentation of 3D medical images23
A confidence-based filtering mechanism, leveraging a feature extraction module combined with a confidence prediction module to quantitatively assess and filter out low-quality pseudo-labels is developed, thereby minimizing the detrimental effects of noisy pseudo-labels on the model's optimization.
- NTIRE 2024 Challenge on Stereo Image Super-Resolution: Methods and Results23
This paper summarizes the 3rd NTIRE challenge on stereo image super-resolution with a focus on new solutions and results and establishes a new benchmark for stereo image SR.
- PFT-SSR: Parallax Fusion Transformer for Stereo Image Super-Resolution23
This letter proposes a novel Transformer-based parallax fusion module called Parallax Fusion Transformer (PFT), and adopted the Swin Transformer as the backbone for feature extraction and SR reconstruction to form a pure Transformer architecture called PFT-SSR.
- 23
- Walking with MIND23
A novel Mental Imagery eNhanceD (MIND) module for the embodied agent, as well as a relevant deep reinforcement framework for training, which proves that the agent with the MIND module is superior to its counterparts not only in EQA performance but in many other aspects such as route planning, behavioral interpretation, and the ability to generalize from a few examples.
- Distinct neural patterns enable grasp types decoding in monkey dorsal premotor cortex23
This study is the first to specify the neuronal population encoding of grasp during the time course of grasp, and demonstrates high grasp decoding performance in PMd, suggesting that PMd may play an important role in generation and maintenance of grasp action and may be a suitable locus for brain-machine interface applications.
- Digital Watermarking Technology for AI-Generated Images: A Survey22
This survey provides a comprehensive review of three core areas: the evolution of image generation technologies, highlighting key milestones such as the transition from GANs to diffusion models; traditional and state-of-the-art digital image watermarking algorithms, encompassing spatial domain, transform domain, and deep learning-based approaches.
- High-Frequency Modulated Transformer for Multi-Contrast MRI Super-Resolution21
A lightweight and accurate High-frequency Modulated Transformer for multi-contrast MRI super-resolution by employing an enhancement module to enhance and amplify the high-frequency priors in the reference and target modalities and a novel cross-attention mechanism to fuse the high-frequency enhanced features with the global features sequentially.
- SILMM: Self-Improving Large Multimodal Models for Compositional Text-to-Image Generation21
This work introduces a model-agnostic iterative self-improvement framework (SILMM) that can enable LMMs to provide helpful and scalable self-feedback and optimize text-image alignment via Direct Preference Optimization (DPO).
- 21
- 21
- Hierarchical Phone Recognition with Compositional Phonetics21
A hierarchical model is presented, in which three different entities are explicitly model three different entities in a hierarchical manner: phoneme, phone, and phonological articulatory attributes, which enables the model to recognize phones even they do not appear in the training set.
- Modern instant messaging platform for postoperative follow-up of patients after total joint arthroplasty may reduce re-admission rate20
The instant messaging platform consultation service allows early detection and management of postoperative adverse events, including problems related to medication, wound, and activity, thereby effectively reducing readmission rate.
- 19
- Lightweight Feature De-redundancy and Self-calibration Network for Efficient Image Super-resolution19
An efficient Feature De-redundancy and Self-calibration Super-resolution network (FDSCSR) is proposed to reduce the repetitive feature information extracted by the model and further enhance the efficiency of the model.
- Self-supervised Meta-Prompt Learning with Meta-Gradient Regularization for Few-shot Generalization18
SUPMER leverages self-supervised meta-learning with a diverse set of well-designed meta-training tasks to learn a universal prompt initialization for efficient adaptation using only unlabeled data, and jointly meta-learns a gradient regularization function to transform raw gradients into a domain-generalizable direction, thus alleviating the problem of overfitting.
- Comparing the Max and Noisy-Or Pooling Functions in Multiple Instance Learning for Weakly Supervised Sequence Learning Tasks18
This paper compares the "max" and "noisy-or" pooling functions on a speech recognition task and a sound event detection task and finds that max pooling is able to localize phonemes and sound events, while noisy-or pooling fails.
- Multiple Instance Deep Learning for Weakly Supervised Small-Footprint Audio Event Detection18
This paper proposes a small-footprint multiple instance learning (MIL) framework for multi-class AED using weakly annotated labels and shows that audio embeddings extracted by the convolutional neural networks significantly boost the performance of all MIL models.
- An efficient personalized federated learning approach in heterogeneous environments: a reinforcement learning perspective17
This work proposes FedPRL, a novel approach to personalized federated learning designed specifically for heterogeneous environments that effectively resolves the challenges of data and system heterogeneity, enhancing the efficiency and model performance of personalized federated learning methods in heterogeneous environments, and outperforming state-of theart methods in terms of model accuracy and training efficiency.
- Dilated Context Integrated Network with Cross-Modal Consensus for Temporal Emotion Localization in Videos17
A cross-modal consensus learning paradigm is introduced, which leverages the inherent semantic consensus between the aligned video and subtitle to achieve weakly-supervised learning on temporal emotion localization.
- An Effective Single-Image Super-Resolution Model Using Squeeze-and-Excitation Networks17
An effective deep residual network based on squeeze-and-excitation blocks (SEBlock) to reconstruct high-resolution (HR) image from low- resolution (LR) image and get finer texture details.
- Eventness: Object Detection on Spectrograms for Temporal Localization of Audio Events16
The concept of Eventness for audio event detection is introduced, which can, in part, be thought of as an analogue to Objectness from computer vision, which has comparable results with a state-of-the-art baseline and is more robust on minority events.
- Benchmarking Multimodal CoT Reward Model Stepwise by Visual Program15
Experimental results demonstrate that SVIP-Reward improves MLLM performance across training and inference-time scaling, yielding better results on benchmarks while reducing hallucinations and enhancing reasoning ability.
- 15
- 15
- 15
- Multiple Degradation and Reconstruction Network for Single Image Denoising via Knowledge Distillation15
The degradation mechanism of the noisy image is explored and a lightweight Multiple Degradation and Reconstruction Network (MDRN) is proposed to progressively remove noise to enable MDRN to learn richer and more accurate features from heterogeneous models, which make it possible to reconstruct higher-quality denoised images under extreme conditions.
- De-fine: De composing and Re fin ing Visual Programs with Auto-Feedback14
Inspired by benders decomposition, this work introduces De-fine, a training-free framework that automatically decomposes complex tasks into simpler subtasks and refines programs through auto-feedback, which can improve logical reasoning performance by integrating the strengths of multiple models.
- Mastering Collaborative Multi-Modal Data Selection: A Focus on Informativeness, Uniqueness, and Representativeness12
A collaborative framework, DataTailor, is proposed, which leverages three key principles,formativeness, uniqueness, and representativeness, for effective data selection and achieves 101.3% of the performance of full-data fine-tuning with only 15% of the data, significantly reducing computational costs while maintaining superior results.
- 12
- 12
- Fact :Teaching MLLMs with Faithful, Concise and Transferable Rationales12
Fact is introduced, a novel paradigm designed to generate multimodal rationales that are faithful, concise, and transferable for teaching MLLMs and filter rationales that can be transferred to end-to-end paradigms from programming paradigms to guarantee transferability.
- 12
- Janus-Pro-R1: Advancing Collaborative Visual Comprehension and Generation via Reinforcement Learning11
This paper proposes to enable the collaborative co-evolution of visual comprehension and generation, advancing image generation into an iterative introspective process and introduces a two-stage training approach: supervised fine-tuning teaches the MLLM with the foundational ability to generate genuine CoT for visual generation, while reinforcement learning activates its full potential via an exploration-exploitation trade-off.
- FormFactory: An Interactive Benchmarking Suite for Multimodal Form-Filling Agents11
A comprehensive evaluation of state-of-the-art MLLMs is conducted and it is observed that no model surpasses 5% accuracy, underscoring the inherent difficulty of the task and revealing significant limitations in current models' visual layout reasoning and field-value alignment abilities.
- Toward Complex-query Referring Image Segmentation: A Novel Benchmark11
A new RIS benchmark with complex queries, namely RIS-CQ is proposed, of high quality and large scale, which challenges the existing RIS with enriched, specific, and informative queries, and enables a more realistic scenario of RIS research.
- 11
- 10
- What Limits Virtual Agent Application? OmniBench: A Scalable Multi-Dimensional Benchmark for Essential Virtual Agent Capabilities10
OmniBench is introduced, a self-generating, cross-platform, graph-based benchmark with an automated pipeline for synthesizing tasks of controllable complexity through subtask composition and OmniEval, a multidimensional evaluation framework that includes subtask-level evaluation, graph-based metrics, and comprehensive tests across 10 capabilities.
- 10
- 10
- Fast MRI Reconstruction via Edge Attention10
An efficient Edge Prediction Network to directly predict accurate edges from the blurred image and a novel Edge Attention Module to guide the image reconstruction utilizing the extracted edge priors, as inspired by the popular self-attention mechanism.
- 10
- Learning 4D Panoptic Scene Graph Generation from Rich 2D Visual Scene9
A 4D Large Language Model (4D-LLM) integrated with a 3D mask decoder for end-to-end generation of 4D-PSG and a 2D-to-4D visual scene transfer learning framework, where a spatial-temporal scene transcending strategy effectively transfers dimension-invariant features from abundant 2D SG annotations to 4D scenes, effectively compensating for data scarcity in 4D-PSG.
- WiseEdit: Benchmarking Cognition- and Creativity-Informed Image Editing9
WiseEdit is introduced, a knowledge-intensive benchmark for comprehensive evaluation of cognition- and creativity-informed image editing, featuring deep task depth and broad knowledge breadth and incorporating three fundamental types of knowledge: Declarative, Procedural, and Metacognitive knowledge.
- DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning9
DAMID-X is introduced, the first dataset to pair AI-generated videos with detailed defect-level, temporal-spatial annotations and written rationales, and DAVID-XR1, a video-language model designed to deliver an interpretable chain of visual reasoning-including defect categorization, temporal-spatial localization, and natural language explanations.
- Unsupervised Domain Adaptation for Video Object Grounding with Cascaded Debiasing Learning9
This paper addresses the Unsupervised Domain Adaptation (UDA) for the dense frame prediction task - Video Object Grounding and measures Grounding Coordinate Offset based on the semantic distance of the model's prediction across domains, based on which it alleviate source-domain bias in the target model through adversarial learning.
- Deep residual refining based pseudo‐multi‐frame network for effective single image super‐resolution9
A deep residual refining based pseudo-multi-frame network for efficient SISR that could make full use of available redundant information possessed in hierarchical layers and can outperform most of the state-of-the-art methods with acceptable less parameters.
- Iris: Breaking GUI Complexity with Adaptive Focus and Self-Refining8
Iris is introduced, a foundational visual agent that addresses challenges in visual perception through two key innovations: Information-Sensitive Cropping (ISC) and Self-Refining Dual Learning (SRDL).
- 8
- 7
- WEAVE: Unleashing and Benchmarking the In-context Interleaved Comprehension and Generation7
The first suite for in-context interleaved cross-modality comprehension and generation for multi-modal community is presented, and it is believed WEAVE provides a view and foundation for studying in-context interleaved comprehension and generation for multi-modal community.
- 7
- I3: I ntent- I ntrospective Retrieval Conditioned on I nstructions7
I3 is introduced, a unified retrieval system that performs Intent-Introspective retrieval across various tasks, conditioned on Instructions without any task-specific training, embodying two key designs: progressive structure pruning and drawback extrapolation-based data refinement.
- Separation and Recovery of Refined Si from Al–Si Melt by Modified Czochralski Method7
Results demonstrate that the modified Czochralski method for the alloy system is an effective way to enrich and separate refined silicon from the Al–Si melt, which provide a potential and clean production of solar grade silicon (SoG-Si) for the future industrial application.
- 7
- Evolving Generalist Virtual Agents with Generative and Associative Memory6
CA3Mem, a framework inspired by the human hippocampus that organizes experiences into a structured memory graph, features two key innovations: a generative memory recombination mechanism that synthesizes novel solutions to drive agent evolution and an associative retrieval algorithm that employs spreading activation to recall a comprehensive and contextually-aware set of experiences.
- Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs6
EvoSynth is an autonomous multi-agent framework that shifts the optimization space from prompts to executable code, and features a code-level self-correction loop, allowing it to iteratively rewrite the code-based algorithm in response to target-model feedback and failed attempts.
- 6
- 6
- 6
- BOSS: Bottom-up Cross-modal Semantic Composition with Hybrid Counterfactual Training for Robust Content-based Image Retrieval6
A novel Bottom-up crOss-modal Semantic compoSition (BOSS) with Hybrid Counterfactual Training framework, which sheds new light on the CIR task by studying it from two previously overlooked perspectives.
- Consistent and Invariant Generalization Learning for Short-video Misinformation Detection5
A new DOmain generalization model via ConsisTency and invariance learning for shORt-video misinformation detection (named DOCTOR), which contains two characteristic modules: the cross-modal feature interpolation to map multiple modalities into a shared space and the interpolation distillation to synchronize multi-modal learning.
- Expression, purification, and crystal structure of mpox virus A41 protein5
Preliminary structural and functional studies of A41 protein from mpox virus will help to better understand its role in chemokine subversion, and contribute to the knowledge to viral chemokine binding proteins.
- Weakly-supervised Disentanglement Network for Video Fingerspelling Detection5
This paper proposes the Weakly-supervised Disentanglement Network, namely WED, that requires no additional knowledge, and better exploits the video-sentence weak supervisions, contributing to the state-of-the-art performance for fingerspelling detection and recognition.
- Progressive Back-Traced Dehazing Network Based on Multi-Resolution Recurrent Reconstruction5
An effective progressive back-traced dehazing network based on multi-resolution recurrent reconstruction strategies that could restore satisfactory high-frequency textures and high-fidelity colors is proposed.
- 4
- Robust Modality-Incomplete Anomaly Detection: A Modality-Instructive Framework with Benchmark4
RADAR, a robust two-stage Robust modAlity-instructive fusing & Detecting frAmewoRk, is proposed, which integrates a Modality-Incomplete Instruction mechanism-guiding the multimodal Transformer to focus more on available modal info, and a Double-Pseudo Hybrid Module to highlight unique modality combinations and reduce overfitting.
- Chart-HQA: A Benchmark for Hypothetical Question Answering in Charts4
A novel Chart Hypothetical Question Answering (HQA) task is introduced, which imposes assumptions on the same question to compel models to engage in counterfactual reasoning based on the chart content, and HAI, a human-AI interactive data synthesis approach that leverages the efficient text-editing capabilities of LLMs alongside human expert knowledge to generate diverse and high-quality HQA data at a low cost.
- DEMON24: ACM MM24 Demonstrative Instruction Following Challenge4
The DEMON Challenge aims to assess the ability of models and systems to comprehend demonstrative instructions consisting of multiple, interleaved, and multimodal context that demonstrate the required information to complete a task.
- B-mode ultrasound-based CAD by learning using privileged information with dual-level missing modality completion4
A novel multi-view LUPI algorithm with Dual-Level Modality Completion, named DLMC-LUPI, is proposed to improve the performance of BUS-based CAD and can effectively improve the performance of single-modal BUS-based CAD.
- 4
- 4
- 4
- 4
- 4
- 4
- A Light-Weight Multimodal Framework for Improved Environmental Audio Tagging4
Experiments on the audio tagging task of the DCASE 2017 challenge show that the incorporation of video information improves a strong baseline audio tagging system by 5.3% in terms of F1 score.
- Multiple Instance Deep Learning for Weakly Supervised Audio Event Detection.4
It is shown that audio embeddings extracted by the convolutional neural networks significantly boost the performance of all MIL models, and this framework reduces the model complexity of the AED system and is suitable for applications where computational resources are limited.
- 3
- 3
- 3
- EvolvedGRPO: Unlocking Reasoning in LVLMs via Progressive Instruction Evolution3
A progressive instruction evolution framework to gradually generate more complex questions via editing instructions in an adversarial way, progressively aligned with the model’s evolving capabilities, which achieves state-of-the-art performance among open-source RL models on multi-modal reasoning tasks, even approaching the closed-source GPT-4o in reasoning capabilities.
- 3
- LASER: Tuning-Free LLM-Driven Attention Control for Efficient Text-conditioned Image-to-Animation3
This paper introduces LASER, a tuning-free LLM-driven attention control framework that follows a progressive process: LLM planning, feature-attention injection, and stable animation generation, and proposes a Text-conditioned Image-to-Animation Benchmark to validate the effectiveness and efficacy of LASER.
- Boosting Private Domain Understanding of Efficient MLLMs: A Tuning-free, Adaptive, Universal Prompt Optimization Framework3
This paper focuses on the efficient adaptation of EMLLMs to private domains, specifically in two areas: how to reduce data requirements, and how to avoid parameter fine-tuning.
- 3
- 3
- Arcadia: Toward a Full-Lifecycle Framework for Embodied Lifelong Learning2
Arcadia delivers consistent gains on navigation and manipulation benchmarks and transfers robustly to physical robots, indicating that a tightly coupled lifecycle: continuous real-world data acquisition, generative simulation update, and shared-representation learning, supports lifelong improvement and end-to-end generalization.
- 2
- Fast Thinking for Large Language Models2
GainRouter is proposed, a lightweight routing mechanism that adaptively switches between fast codebook guided inference and slow explicit reasoning, thereby suppressing overthinking and reducing unnecessary token generation in large language models.
- 2
- 2
- 2
- 2
- 1
- 1
- Counterfactual Evolution of Multimodal Datasets via Visual Programming1
Experiments show that SCOPE improves reasoning performance, exposes model blind spots, and enhances visual dialog capabilities, and is proposed to support training, a curriculum learning strategy that aligns model capacity with sample difficulty.
- A lightweight self-ensemble feedback recurrent network for fast MRI reconstruction1
A novel Self-Ensemble Feedback Recurrent Network (SEFRN) for fast MRI reconstruction inspired by recursive learning and ensemble learning strategies is proposed and achieves competitive results and strikes a good balance between the size, complexity, and performance of the model.
- 1
- 1
- Reasoning Makes Good Annotators : An Automatic Task-specific Rules Distilling Framework for Low-resource Relation Extraction1
This work guides the pre-trained language model to reason rules as experts and compose them into robust compound rules for data labeling, which could continuously enrich the rules set to power the labeling ability by discovering reliable model-labeled data for distinguishable rules generation.
- 1
- 1
- The Call of Social Capital in Pinduoduo Withdrawal Activities from The Perspective of Social Network Theory1
The investigation of social network calls in withdrawal activities found that the more people at the top of the network and the greater the difference in social networks, the more the possibility of successful withdrawal.
- 1
- 1
- 1
- 1
- 1
- Symmetrical Uncertainty Based Neural Selection in Motor Cortex Brain-Machine Interfaces1
A novel algorithm based on symmetrical uncertainty that optimally selects neurons to decode kinetic variables is proposed that both relevance and redundancy are considered and it is shown that the neurons selected by the method are less redundant.
- –
- –
- Unified multi-prototype network with pretrained swin transformer for visual and audio open set recognition–
This work proposes a novel framework called multi-prototype network with Pretrained Swin Transformer (PST-MPN) for OSR, developed based on a novel multi-prototype learning mechanism combined with hybrid distance that significantly outperforms other baseline methods and obtains state-of-the-art performance for OSR on both visual and audio tasks.
- –
- Trusted Multi‐View Graph Convolutional Network With Cross‐View GraphMAE for Alzheimer's Disease Classification–
Results indicate that CGM‐TMVGCN effectively integrates LOFC and HOFC features for reliable AD classification, outperforming competing multi‐view GCN and graph self‐supervised learning methods.
- Generation-Augmented Video Corpus Moment Retrieval–
This work proposes Video-GAR, a novel framework that reframes the conventional retrieval task from superficial matching to generative understanding, positing that the capability for query reconstruction evidences deep semantic comprehension.
- –
- A transformer fault identification method based on an improved learning numerical spiking neural P system–
The experimental results demonstrate that learning numerical spiking neural P system based on improved Bayesian inference (Bi-LNSN P system) proposed can realize the improvement of transformer fault recognition accuracy.
- In-band full-duplex technology in VDES: a hardware implementation for high isolation in the analogue domain–
In this paper, the in-band full-duplex technology is applied to the VHF Data Exchange System (VDES) to eliminate the self-interference signal and improve the isolation between transceiver links in the 160 MHz frequency band.
- –
- Ψ-Net: Triple-Branch Network with Cross-Branch Alternately Updated Fusion for Diagnosis of Bicuspid Aortic Valve Via Dual-View Echocardiography–
A Triple-Branch Network (named Ψ-Net) is proposed as a Computer-Aided Diagnosis (CAD) model for BAV based on the paired TTE images of aortic valve, suggesting its effectiveness in the diagnosis of BAV.
- Traffic Image Restoration under Adverse Weather via Frequency-Aware Mamba–
Frequency-Aware Mamba is proposed, a novel framework that integrates frequency guidance with sequence modeling for efficient image restoration and a novel Adaptive Frequency Scanning Mechanism for the Mamba, which enables the Mamba to achieve frequency-domain scanning across distinct subgraphs, thereby fully leveraging the texture distribution characteristics inherent in subgraph structures.
- –
- Uncertainty‐Aware Graph Self‐Training for Autism Spectrum Disorder Classification in Multiple Centers–
Experimental results verified the effectiveness of the proposed uncertainty‐aware graph self‐training (UA‐GST) by extending graph self‐training to a new situation, in which the labeled data come from one imaging center and the unlabeled data come from several other imaging centers.
- The 2nd International Workshop on Deep Multi-modal Generation and Retrieval–
This workshop aims to encourage more explorations in Deep Multimodal Generation and Retrieval, providing a platform for researchers to share insights and advancements in this rapidly evolving domain.
- –
- –
- Research on Evaluation Index System of Teacher Satisfaction for Chinese Digital Educational Platforms–
This study aims to comprehensively investigate and analyze the factors influencing teacher satisfaction towards these platforms, exploring teachers' attitudes, usage patterns, and experiential feedback across different platforms.
- –
- –
- –
- –
- Clinical Test on Treatment of Paraplegia by Supreme Rehabilitation Act after Spinal Cord Injured–
It is concluded that supreme rehabilitation can improve patients’ motor function, sensory function and self-care ability significantly in the rehabilitation of paraplegia after spinal cord injured.
- Personal Credit Assessment of Student Loans Based on Improved Analytic Hierarchy Process–
The results show that the proposed method is reasonable and effective that can provides a simpler and more accurate basis of decision for student loans.
- –
- The Research of Face Recognition Method Based on Sparse Representation and Feature Selection–
This paper proposes a fast decomposition gradient projection algorithm in solving sparse representation (FDGP), which converts large-scale optimization problems small bounded constrained quadratic programming problem to solve and saves memory consumption.
- Laboratory Management of Animal and Veterinary Profession in Vocational Institutions–
The scientific, standardized and systematic management of laboratory was promoted so as to completely promote the students′ practical skills.
- Development and Application of Professional Database of Multi-function Gastric Cancer Clinical Pathology Data–
The multi-function gastric cancer clinical pathology database that was set up by virtue of Filemaker pro 8.5 database software is a practical tool for saving, analyzing and summarizing the Gastric cancer data.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.