Posters

DiffPano generates scalable, consistent, and diverse panoramic images from text descriptions and camera poses using a novel spherical epipolar-aware diffusion model.

DIFFNORM boosts non-autoregressive speech-to-speech translation by normalizing speech data with a diffusion model and classifier-free guidance, achieving significant quality improvements.

DiffHammer unveils weaknesses in diffusion-based adversarial defenses by introducing a novel attack bypassing existing evaluation limitations, leading to more robust security solutions.

Tighter differential privacy (RDP) guarantees for DP-SGD with fixed-size minibatches are achieved, improving private deep learning model training.

Differentially private set representations achieve optimal privacy-utility tradeoffs with exponentially smaller error than prior histogram methods.

This paper presents DP-Nash-VI, a novel algorithm ensuring trajectory-wise privacy in multi-agent reinforcement learning, achieving near-optimal regret bounds under both joint and local differential p…

This paper presents new, nearly optimal differentially private algorithms for handling sparse gradients, significantly improving efficiency and scalability in large embedding models.

This paper introduces a novel differentially private graph diffusion framework ensuring edge-level privacy, significantly improving utility-privacy trade-offs for personalized PageRank computation.

First differentially private algorithm for testing equivalence between continuous distributions, enabling privacy-preserving comparisons of sensitive data.

Differentially private scalable kernel learning is achieved via a novel DP K-means Nyström method, enabling efficient and accurate model training for general kernels while safeguarding privacy.

This research introduces a novel plug-and-play module that efficiently integrates prior partial order constraints into differentiable structure learning, significantly improving structure recovery qua…

Quantum algorithm achieves super-quadratic speedup for large-scale linear control, offering a novel approach to address the computational challenges of optimizing complex dynamical systems.

Differentiable Modal Synthesis (DMSP) achieves superior accuracy in simulating the dynamic behavior of nonlinear strings, enabling realistic sound and motion synthesis guided by physical laws.

Diffeomorphic interpolation boosts topological optimization by transforming sparse gradients into smooth vector fields, enabling efficient large-scale point cloud optimization and black-box autoencode…

DiffCut, a novel unsupervised zero-shot semantic segmentation method, leverages diffusion UNet features and recursive normalized cuts to achieve state-of-the-art performance.

Boost classifier robustness with DiffAug, a novel diffusion-based augmentation method! One forward and reverse diffusion step enhances robustness against covariate shifts, adversarial examples, and o…

Diff-eRank: A novel rank-based metric assessing LLMs’ efficiency in eliminating redundant information during training, showing improved correlation with model size and performance.

DI-MaskDINO: Novel model significantly boosts object detection & instance segmentation accuracy by addressing performance imbalance using a De-Imbalance module and Balance-Aware Tokens Optimization.

Decoupled-Head Attention (DHA) drastically cuts LLM inference costs by adaptively sharing key/value heads, achieving 97.6% of original performance with only 0.25% pre-training.

DG-SLAM achieves robust real-time visual SLAM in dynamic scenes using 3D Gaussian splatting and a novel hybrid pose optimization, significantly outperforming existing methods.