Posters

MERIT-IRL: First in-trajectory IRL framework learns reward & policy incrementally from ongoing trajectories, guaranteeing sub-linear regret.

In-N-Out: Lifting 2D Diffusion Priors for 3D Object Removal via Tuning-Free Latents Alignment enhances 3D scene reconstruction by aligning 2D diffusion model latents for consistent multi-view inpainti…

CONTEXTSSL: A novel self-supervised learning algorithm that adapts to task-specific symmetries by using context, achieving significant performance gains over existing methods.

Transformers learn contextual information for generalization to unseen examples and tasks, even with limited training data, converging linearly to a global minimum.

This paper introduces inner and momentum optimization to enhance the state vector for in-context learning, improving performance and scalability in LLMs.

Linear Transformer Blocks (LTBs) achieve near-optimal in-context learning (ICL) for linear regression by effectively implementing one-step gradient descent with learnable initialization, a significant…

This research leverages causal intervention to identify and utilize genuine label correlations in multi-label image recognition, mitigating contextual bias for improved accuracy.

Active Viewpoint Selection (AVS) significantly improves viewpoint-independent object-centric representations by actively selecting the most informative viewpoints for each scene, leading to better seg…

Researchers significantly boosted the efficiency and quality of rectified flow, a method for generating samples from diffusion models, by introducing novel training techniques that surpass state-of-th…

Deep Fourier Shifting boosts small image restoration networks by using an information-lossless Fourier cycling shift operator, improving performance across various low-level tasks while reducing compu…

TPNet boosts temporal link prediction accuracy and efficiency by unifying relative encodings via temporal walk matrices and using random feature propagation.

Data Debiasing with Datamodels (D3M) efficiently improves machine learning model robustness by identifying and removing specific training examples that disproportionately harm minority groups’ accurac…

Gated Sparse Autoencoders (GSAEs) achieve Pareto improvement over baseline SAEs for unsupervised feature discovery in language models, resolving the shrinkage bias of L1 penalty by separating feature …

Anchored Confidence (AnCon) significantly improves self-training under distribution shifts by using a temporal ensemble to smooth noisy pseudo-labels, achieving 8-16% performance gains without computa…

Boosting 3D point cloud recognition robustness, Frequency Adversarial Training (FAT) leverages frequency-domain adversarial examples to improve model resilience against corruptions, achieving state-of…

Boosting Neural ODE training, Temporal Adaptive Batch Normalization (TA-BN) resolves traditional Batch Normalization’s limitations by providing a continuous-time counterpart, enabling deeper networks …

Boosting neural network surface processing: Using principal curvatures as input significantly improves segmentation and classification accuracy while reducing computational overhead.

Accelerate Gaussian process hyperparameter optimization by up to 72x using novel linear system solver techniques.

SignCL, a novel contrastive learning strategy, significantly boosts gloss-free sign language translation by mitigating representation density, achieving substantial performance gains.

EpoD, a novel dynamic graph learning model, significantly improves generalization via a self-prompted learning mechanism for environment inference and a structural causal model utilizing dynamic subgr…