5494 papers
The paper introduces ScribeTokens, a fixed-vocabulary tokenization method for digital ink that decomposes pen movements into unit pixel steps, demonstrating superior performance over vector representations in both handwritten text generation and recognition, particularly when enhanced by a novel next-ink-token prediction pretraining strategy.
This paper introduces GaussDerResNets, a novel deep learning architecture that combines scale-covariant Gaussian derivative layers with residual skip connections to achieve provable scale invariance and superior generalization to unseen image scales across multiple datasets while maintaining high accuracy and computational efficiency.
By probing LVLMs with a synthetic directed graph dataset, this study reveals that while node and structural information are linearly encoded early in the vision encoder, edge representations emerge only later in the language model's text tokens, explaining the models' persistent struggles with relational understanding.
This paper introduces a multimodal-prior-guided importance sampling framework for hierarchical 3D Gaussian Splatting that fuses photometric, semantic, and geometric cues to strategically refine sparse-view novel view synthesis, thereby achieving state-of-the-art reconstruction quality while mitigating overfitting and noise.
SIGMark is a scalable, distortion-free in-generation watermarking framework for video diffusion models that enables blind extraction and robustness against temporal disturbances through Global Frame-wise PseudoRandom Coding and a Segment Group-Ordering module tailored for causal 3D VAEs.
The paper proposes SemanticDialect, a semantic-aware mixed-format quantization framework for Video Diffusion Transformers that utilizes block-wise format selection, activation decomposition with residual correction, and semantic token grouping to achieve high-quality video generation with significantly reduced memory and compute costs.
StegaFFD is a privacy-preserving face forgery detection framework that embeds facial images into natural cover images using steganography to avoid suspicion, while employing specialized decomposition, attention, and alignment mechanisms to accurately detect forgeries within the steganographic domain despite semantic interference.
This paper introduces LLandMark, a modular multi-agent framework that enhances multimodal video retrieval for complex, real-world queries by integrating specialized agents for landmark detection, LLM-driven image generation, and OCR refinement to achieve culturally grounded and explainable results, particularly for Vietnamese scenes.
MVD-HuGaS is a novel framework that achieves state-of-the-art free-view 3D human rendering from a single image by leveraging a fine-tuned multi-view diffusion model to generate consistent multi-view images, an alignment module for joint Gaussian and pose optimization, and a depth-based facial distortion mitigation module to ensure high-fidelity reconstruction.
This paper introduces 3D-DRES, a new task for fine-grained phrase-to-3D-instance mapping, supported by the DetailRefer dataset with pioneering phrase-instance annotations and the DetailBase baseline, which together enhance 3D vision-language understanding and improve performance on traditional 3D-RES benchmarks.
The paper proposes ProGIC, a progressive and lightweight generative image compression codec based on residual vector quantization and a compact backbone, which achieves significant bitrate savings, faster encoding/decoding speeds, and flexible progressive transmission compared to existing methods.
This paper introduces the Harmonic Beltrami Signature Network (HBSN), a novel deep learning architecture that efficiently computes rotation-, scale-, and translation-invariant shape representations to serve as a versatile shape prior module, thereby enhancing the performance of existing computer vision segmentation models.
This paper presents Articulation in Motion (AiM), a prior-free framework that leverages a dual-Gaussian scene representation and sequential RANSAC to automatically segment articulated objects into rigid parts, estimate their kinematics, and reconstruct interactive 3D replicas from a single static scan and an interaction video without requiring prior knowledge of the number of parts.
HDINO is a concise and efficient open-vocabulary detector that eliminates reliance on manually curated datasets and resource-intensive feature extraction by employing a two-stage training strategy with a One-to-Many Semantic Alignment Mechanism and Difficulty Weighted Classification Loss to achieve state-of-the-art performance on COCO with significantly fewer training images than existing methods.
GloPath is a scalable, entity-centric foundation model trained on over one million glomeruli that outperforms state-of-the-art methods in diverse lesion assessment tasks and uncovers significant clinicopathological associations, advancing the clinical translation of AI in renal pathology.
TC-Padé is a novel feature prediction framework that leverages Trajectory-Consistent Padé approximation with adaptive coefficient modulation and step-aware strategies to significantly accelerate diffusion models in low-step regimes while maintaining high generation quality and overcoming the trajectory drift limitations of existing polynomial-based methods.
This paper proposes a lightweight semi-supervised learning framework that introduces a stochastic Proportion Loss, adapted from learning from label proportions, to align model predictions with global class distributions and effectively mitigate class imbalance issues in SSL.
This paper proposes an efficient semi-supervised method for few-shot adaptation of vision-language models in medical imaging that leverages unlabeled data to propagate text-informed pseudo-labels, thereby reducing annotation requirements by over 50% while addressing class imbalance challenges.
This paper proposes a framework combining a foundation-model-based anomaly synthesis pipeline (FMAS) and a Wavelet Domain Attention Module (WDAM) to generate realistic synthetic anomalies and enhance feature extraction, significantly improving industrial anomaly detection performance on benchmark datasets without requiring fine-tuning.
TagaVLM is an end-to-end framework that enhances Vision-Language Navigation by explicitly injecting topological structures into the VLM backbone via Spatial Topology Aware Residual Attention and Interleaved Navigation Prompts, achieving state-of-the-art performance on the R2R benchmark and demonstrating that targeted architectural improvements on smaller models can outperform brute-force scaling for embodied spatial reasoning.