Posters

µP²: Layerwise perturbation scaling in SAM enables hyperparameter transfer and improved generalization in large models.

ZOPP: A groundbreaking framework for zero-shot offboard panoptic perception in autonomous driving, enabling high-quality 3D scene understanding without human labeling.

Zipfian Whitening: Weighting PCA whitening by word frequency dramatically improves NLP task performance, surpassing established baselines and providing a theoretical framework for existing methods.

ZipCache: Efficient KV cache quantization for LLMs using salient token identification, achieving 4.98x compression with minimal accuracy loss!

Zeroth-Order Diffusion Monte Carlo (ZOD-MC) efficiently samples from non-log-concave distributions using only zeroth-order queries, overcoming metastability issues and outperforming state-of-the-art s…

ZeroMark revolutionizes dataset ownership verification by enabling copyright protection without exposing watermarks, leveraging the intrinsic properties of DNNs trained on watermarked data.

Zero-to-Hero enhances zero-shot novel view synthesis by cleverly filtering attention maps during inference, achieving significantly higher fidelity and realism without retraining.

Zero-shot Neural ODEs enable autonomous systems to rapidly adapt to unseen scenarios by learning a space of dynamical systems spanned by neural ODE basis functions, achieving efficient online adaptati…

Zero-Shot Tokenizer Transfer (ZeTT) detaches language models from their tokenizers via a hypernetwork, enabling efficient on-the-fly tokenizer swapping without retraining, significantly improving LLM …

Zero-shot 3D scene reconstruction from single images is achieved by assembling diverse deep priors from large models, eliminating the need for 3D/2D training data and achieving superior performance.

Zero-shot RL struggles with low-quality data; this paper introduces conservative algorithms that significantly boost performance on such data without sacrificing performance on high-quality data.

MimicBrush: a novel image editing approach using reference imitation for intuitive zero-shot edits.

ZiRa achieves zero-shot generalizable incremental learning for vision-language object detection by using a memory-efficient dual-branch architecture and zero-interference loss, significantly boosting …

Zero-shot Event-Intensity Asymmetric Stereo (ZEST) uses visual prompting and monocular cues to achieve robust 3D perception without event-specific training, outperforming existing methods.

Diffusion models benefit from contrastive training, improving sample quality and speed by addressing poor denoiser estimation in out-of-distribution regions.

Contrastive learning is reframed as a distribution alignment problem, leading to a flexible framework (GCA) that improves representation learning with unbalanced optimal transport.

YOUDREAM generates anatomically consistent, high-quality 3D animal models from text and 2D pose priors, pushing creative boundaries in text-to-3D generation.

YOLA: A novel framework for object detection in low-light conditions, achieving significant improvements by learning illumination-invariant features through a novel module.

Self-supervised learning’s reliance on complex data augmentations is challenged; a large-scale study shows comparable performance using only cropping, suggesting dataset size is more important than au…

YOLOv10: Real-time object detection achieves state-of-the-art speed and accuracy by eliminating NMS post-processing and holistically optimizing model architecture for efficiency and accuracy.