Posters

ActFusion: a unified diffusion model achieving state-of-the-art performance in both action segmentation and anticipation by jointly learning visible and invisible parts of video sequences.

ActAnywhere, a novel video diffusion model, seamlessly integrates foreground subjects into new backgrounds by generating realistic video backgrounds tailored to subject motion, significantly reducing …

First tractable algorithm achieves minimax optimal regret in average-reward MDPs, solving a major computational challenge in reinforcement learning.

D-AdaST: A novel distributed adaptive minimax optimization method achieves near-optimal convergence by tracking stepsizes, solving the inconsistency problem hindering existing adaptive methods.

A novel parameter-free decentralized optimization algorithm achieves linear convergence for strongly convex, smooth objectives, eliminating the need for hyperparameter tuning and improving scalability…

Certifying neural network robustness across diverse domains, this paper introduces knowledge continuity—a novel framework ensuring model stability independent of input type, norms, and distribution.

Cert-LSVI-UCB achieves constant regret in RL with linear function approximation, even under model misspecification, using a novel certified estimator.

Constrained Markov Decision Processes (CMDPs) get an improved sample complexity bound of Õ(1/ε) via a new algorithm, surpassing the existing O(1/ε²) bound.

This paper introduces a computationally efficient method to approximate the optimal accuracy-fairness trade-off curve for various datasets, providing rigorous statistical guarantees and quantifying un…

This paper introduces a novel framework for evaluating the robustness of machine learning models when the true data distribution is only partially known. It defines a new risk measure (‘identifiable r…

ACFun: A novel facial stylization method fusing abstract & concrete features for high-quality, artistically pleasing results from only one style & one face image.

Researchers enhanced transformer models for inertial pose estimation by introducing a Sequence Structure Module, leveraging inherent fixed-length sequence structures for improved accuracy and steadine…

LLM compression accuracy hides crucial behavioral changes; use % flips and KL-divergence for better evaluation.

Accelerated gradient-free optimization is achieved using only function value comparisons, significantly improving black-box optimization.

PITOME: a novel token merging method accelerates Transformers by 40-60% while preserving accuracy, prioritizing informative tokens via an energy score.

Space partitioning dramatically speeds up relative entropy coding (REC) for neural compression, achieving 5-15% better bitrates than previous methods.

Chain-of-Sight accelerates multimodal LLM pre-training by ~73% using a multi-scale visual resampling technique and a novel post-pretrain token scaling strategy, achieving comparable or superior perfor…

This paper presents QSI-NMS and BOE-NMS, novel graph theory-based algorithms that significantly speed up non-maximum suppression (NMS) in object detection without significant accuracy loss, and introd…

MCCFVFP, a novel Monte Carlo-based algorithm, accelerates Nash equilibrium convergence in large-scale games by combining CFR’s counterfactual value calculations with fictitious play’s best response st…

Fast imprecise oracles drastically reduce query times in matroid optimization, achieving near-optimal performance with few accurate queries.