7831 papers
MolFM-Lite is a multi-modal machine learning model that improves molecular property prediction by jointly encoding 1D sequences, 2D graphs, and 3D conformer ensembles through cross-attention fusion and FiLM conditioning, achieving significant performance gains over single-modality baselines on MoleculeNet benchmarks.
This paper introduces SimpleOCR, a plug-and-play training strategy that renders text queries directly onto images to force Multimodal Large Language Models to overcome "modality laziness" and genuinely read visual text, achieving significant performance gains on out-of-distribution benchmarks with extreme data efficiency.
This paper demonstrates the feasibility of deploying privacy-preserving, real-time episodic memory question answering on edge devices by utilizing a two-threaded pipeline with Multimodal Large Language Models, achieving competitive accuracy and low latency compared to cloud-based solutions.
MammoWise is a practical, local, multi-model pipeline that leverages open-source Vision Language Models enhanced with few-shot prompting and Retrieval Augmented Generation to generate high-quality mammography reports and perform accurate clinical classifications while ensuring privacy and reproducibility.
This paper proposes Space Syntax-guided Post-training (SSPT), a framework that integrates architectural theory into residential floor plan generation via a non-differentiable oracle and reinforcement learning to enhance public space dominance and functional hierarchy while outperforming distribution-fitted baselines in efficiency and stability.
Pix2Key is a novel composed image retrieval framework that utilizes semantic decomposition and self-supervised visual dictionary learning to represent queries and candidates as open-vocabulary dictionaries, thereby achieving superior intent-aware matching and diversity-aware reranking without relying on supervised triplets.
This paper introduces HARU-Net, a novel Hybrid Attention Residual U-Net architecture that integrates hybrid attention transformers and residual learning to effectively denoise low-dose Cone-Beam Computed Tomography (CBCT) images while preserving critical anatomical edges, outperforming state-of-the-art methods in both image quality metrics and computational efficiency.
DisQ-HNet is a novel, interpretable framework that synthesizes tau-PET images from T1 and FLAIR MRI by employing a Partial Information Decomposition-guided vector-quantized encoder and a Half-UNet decoder to disentangle modality contributions while preserving anatomical details and disease-relevant signals for Alzheimer's disease analysis.
DrivePTS is a progressive learning framework that enhances autonomous driving scene generation by mitigating geometric condition inter-dependencies, enriching semantic context through multi-view hierarchical text descriptions, and improving structural fidelity via frequency-guided loss, thereby achieving state-of-the-art realism and controllability.
SwiftNDC is a fast, general framework that employs a Neural Depth Correction field to generate cross-view consistent depth maps, providing a robust geometric initialization that significantly accelerates 3D Gaussian Splatting for high-fidelity mesh reconstruction and improves novel-view synthesis quality.
The paper proposes Quality-Aware Robust Multi-View Clustering (QARMVC), a novel framework that addresses heterogeneous observation noise by leveraging reconstruction discrepancies to generate instance-level quality scores, which then guide a hierarchical learning strategy to adaptively suppress noise and construct a robust global consensus.
This paper exposes a critical evaluation pitfall where common human preference models are biased toward large guidance scales, leading to inflated scores despite degraded image quality, and proposes a novel guidance-aware evaluation framework (GA-Eval) alongside a new method (TDG) to demonstrate that many recent diffusion guidance improvements are illusory and that simply increasing CFG scales often outperforms them in practice.
GIFSplat introduces a purely feed-forward, iterative refinement framework for 3D Gaussian Splatting from sparse unposed views that distills a frozen diffusion prior into Gaussian-level cues to achieve state-of-the-art reconstruction quality with second-scale inference time, eliminating the need for camera poses or test-time optimization.
This paper introduces Causal Motion Diffusion Models (CMDM), a unified framework that combines a Motion-Language-Aligned Causal VAE with an autoregressive diffusion transformer to achieve high-quality, real-time, and temporally consistent motion generation while overcoming the limitations of existing bidirectional and unstable autoregressive approaches.
BetterScene enhances novel view synthesis for sparse, unconstrained real-world photos by integrating a feed-forward 3D Gaussian Splatting model with a Stable Video Diffusion backbone that is fine-tuned via temporal equivariance regularization and vision foundation model-aligned representations within its VAE module to produce consistent, artifact-free views.
This paper introduces -DPO, a novel Fairness Direct Preference Optimization framework for Large Multimodal Models that mitigates both catastrophic forgetting and data imbalance-induced bias through a new loss function and pairwise preference alignment, achieving state-of-the-art performance in continual learning benchmarks.
The paper presents LoR-LUT, a unified low-rank formulation that generates compact, interpretable 3D lookup tables by combining basis LUTs with low-rank residual corrections to achieve expert-level image retouching with high perceptual fidelity and a sub-megabyte model size.
This paper proposes DP-aware AdaLN-Zero, a sensitivity-aware conditioning mechanism for diffusion transformers that mitigates heavy-tailed gradients caused by heterogeneous contexts, thereby reducing clipping bias and improving utility in differentially private time-series tasks without compromising standard performance.
The paper introduces SATtxt, a vision-language foundation model that distills multi-spectral knowledge into an RGB-only inference framework aligned with instruction-augmented LLMs, thereby achieving superior zero-shot classification, retrieval, and linear probing performance on satellite imagery benchmarks.
This paper presents Coded-E2LF, a purely event-based computational imaging method that utilizes a coded aperture to reconstruct 4-D light fields with pixel-level accuracy from a stationary event-only camera, marking the first demonstration of achieving this without relying on intensity images.