Posters

Predictive Attractor Models (PAM) offer a biologically-plausible, streaming sequence memory architecture that avoids catastrophic forgetting and generates multiple future possibilities.

This paper presents a novel statistical framework for ranking LLMs using pairwise comparisons, accounting for the uncertainty introduced when using an LLM instead of human preferences. The framework …

PAD, a novel visual policy learning framework, unifies image prediction and robot action in a joint denoising process, achieving significant performance improvements in robotic manipulation tasks.

Foundation model OOD performance prediction is reliably achieved via ensemble diversity, especially through random linear head initialization, enabling precise estimations without extensive OOD labels…

Boosting label distribution learning accuracy and efficiency, this research proposes using ternary labels instead of binary labels to predict label distributions, thus enhancing annotation efficiency …

Deep learning algorithms now predict quantum ground state properties with constant sample complexity, regardless of system size, improving upon previous methods.

Predicting RL agent behavior is key for safety and interaction; this study reveals that explicitly planned agents are significantly easier to predict due to their internal plans.

New analysis reveals how reweighted least-squares algorithms for linear diagonal networks achieve favorable performance in high-dimensional settings, improving upon existing theoretical guarantees and…

SpatioTemporal Video Diffusion (STVD) revolutionizes precipitation downscaling using a novel diffusion model, achieving state-of-the-art accuracy in super-resolution while capturing crucial distributi…

Researchers achieved high-accuracy differentially private (DP) models by using a novel DP continual pre-training strategy with only 10% public data, mitigating the performance degradation common in DP…

MGPO: Efficient online RL adaptation via prompt optimization of pre-trained multi-goal transformers.

Pre-trained LLMs surprisingly use Fourier features to perform addition, with MLP layers approximating magnitude and attention layers handling modular arithmetic; this mechanism requires pre-training.

Revolutionizing unordered data compression, this paper introduces autoregressive shuffle coding, achieving state-of-the-art speeds and compression rates on massive datasets.

PS-BAX, a novel Bayesian algorithm execution method using posterior sampling, efficiently selects evaluation points for complex tasks, outperforming existing methods in speed and scalability.

A novel algorithm achieves a 0.385-approximation for submodular maximization under cardinality constraints, combining strong theoretical guarantees with practical query complexity.

Parametric Piecewise Linear Networks (PPLNs) achieve state-of-the-art results in event-based and frame-based computer vision tasks by mimicking biological neural principles.

PowerPM: A foundation model revolutionizing power system analysis by mastering complex ETS data through a novel self-supervised pre-training approach, achieving state-of-the-art performance.

PiMForce: Robust hand pressure estimation using 3D hand posture and sEMG!

Post-hoc model transformations can reverse performance trends, prompting a reevaluation of model selection strategies and suggesting a new ‘post-hoc selection’ method for improved model development.

Position coupling, a novel method, enhances the length generalization ability of arithmetic Transformers by directly embedding task structures into positional encodings. This simple technique enables…