Posters

Risk control boosts early-exit neural networks’ speed and safety by ensuring accurate predictions before exiting early, achieving substantial computational savings across diverse tasks.

Fast Tree-Field Integrators (FTFIs) revolutionize graph processing by enabling polylog-linear time computation for integrating tensor fields on trees, providing significant speedups for various machin…

TRAC: a parameter-free optimizer conquering lifelong RL’s plasticity loss!

Fast T2T: Optimization Consistency Boosts Diffusion-Based Combinatorial Optimization!

Accelerated discrete diffusion model sampling is achieved via novel discrete non-Markov diffusion models (DNDM) with predetermined transition times, enabling a training-free algorithm that significant…

Conditional Conjugate Integrators (CCI) drastically accelerate sampling in iterative refinement models for inverse problems, achieving high-quality results with only a few steps.

This paper introduces a novel approach for fast rates in online convex optimization by exploiting the curvature of feasible sets, achieving logarithmic regret bounds under specific conditions.

This paper presents efficient algorithms for designing cost-optimal proxy experiments to identify causal effects, significantly improving upon prior methods.

Forgetful algorithms are essential for fast last-iterate convergence in learning games; otherwise, even popular methods like OMWU fail.

Accelerate iterative hard thresholding (IHT) up to 73x by safely pruning unnecessary gradient computations without accuracy loss.

Fast Graph Sharpness-Aware Minimization (FGSAM) accelerates few-shot node classification by cleverly combining GNNs and MLPs for efficient, high-performing training.

TRACKSTO4D: Fast & accurate 3D reconstruction from casual videos using 2D point tracks, drastically reducing runtime by up to 95% while matching state-of-the-art accuracy.

Polar codes revolutionize channel simulation, offering scalable, high-performance schemes that significantly outperform existing methods.

Speculative Rejection: A novel algorithm boosts Large Language Model (LLM) alignment by speeding up inference-time alignment by 16-32x!

Streamlined Inference, a novel training-free framework, dramatically reduces the computation and memory costs of video diffusion models without sacrificing quality, enabling high-resolution video gene…

FasMe: a novel meta-learning approach delivers fast and sample-efficient precision matrix estimation, surpassing existing methods in accuracy and speed for small sample datasets.

FashionR2R leverages diffusion models to realistically translate rendered fashion images into photorealistic counterparts, enhancing realism and preserving fine-grained clothing textures.

FairWire tackles structural bias in graph machine learning, proposing a novel fairness regularizer and a fair graph generation framework for unbiased link prediction and graph generation.

FairQueue improves fair text-to-image generation by addressing prompt learning’s quality issues through prompt queuing and attention amplification.

Nash bargaining resolves hypergradient conflicts in fairness-aware meta-learning, boosting model performance and fairness.