Posters

Soft labels, not sophisticated data synthesis, are the key to successful dataset distillation, significantly improving data-efficient learning and challenging existing methods.

This paper introduces DCL-KR and DCL-NN, novel distillation-based collaborative learning algorithms achieving nearly minimax optimal convergence rates in heterogeneous environments without direct data…

Huber loss minimization ensures accurate and robust mean estimation under user-level differential privacy, especially for imbalanced datasets and heavy-tailed distributions.

New nonparametric framework explains ML performance gaps across domains by hierarchically decomposing discrepancies due to covariate and conditional outcome shifts, offering detailed variable-level at…

CONTACCUM: Stable, efficient memory reduction for dense retrievers using dual memory banks, surpassing high-resource baselines.

This paper introduces mSSRM-PGA, achieving globally optimal m-sparse Sharpe ratios, addressing the nonconvexity issue in portfolio optimization through a novel proximal gradient algorithm.

Depth-range-free MVS network using pose embedding achieves robust and accurate 3D reconstruction.

Generative model learns data symmetries for improved efficiency and higher test log-likelihoods.

Researchers developed a lightweight protocol to probe large vision models’ 3D physical understanding by training classifiers on model features for various scene properties (geometry, material, lightin…

This paper introduces semi-structured functional networks (SSFNNs), a novel approach that combines interpretable functional regression models with deep neural networks, achieving both high accuracy an…

This paper introduces a novel full-duplex speech dialogue system based on LLMs, achieving significantly reduced response latency and higher interruption precision compared to half-duplex systems.

This paper introduces a novel framework for bilevel optimization on Riemannian manifolds, providing efficient hypergradient estimation strategies and convergence analysis, with successful applications…

ULTRAQUERY: a groundbreaking foundation model for zero-shot logical query reasoning on any knowledge graph, surpassing existing methods’ limitations.

Flexible Subgraph GNNs, achieving scalability via graph products and coarsening, consistently outperform baselines and adapt to varying subgraph numbers.

Revolutionizing probabilistic inference, PLIA₁ uses tensor operations and FFT to scale integer arithmetic, achieving orders-of-magnitude speedup in inference and learning times.

Transformers learn increasingly complex language patterns sequentially, starting with simpler interactions before mastering higher-order ones.

LLMs dynamically adjust restless multi-armed bandit (RMAB) resource allocation policies in public health via human-language commands.

Contrary to popular belief, simple supervised fine-tuning with strong language models outperforms complex reinforcement learning in aligning large language models, significantly improving efficiency.

CAST: a novel consistency-aware spot-guided Transformer achieves state-of-the-art accuracy and efficiency in point cloud registration.

CoX-LMM unveils a novel concept-based explainability framework for large multimodal models, extracting semantically grounded multimodal concepts to enhance interpretability.