🏢 Carnegie Mellon University

Provably robust diffusion posterior sampling for plug-and-play image reconstruction is achieved via a novel algorithmic framework, DPnP, offering both asymptotic and non-asymptotic performance guarant…

Unsupervised learning predicts protein-nucleic acid binding using contact map prediction, significantly improving aptamer screening via FAFormer, a novel equivariant transformer.

This paper introduces a novel privacy-preserving algorithm for personalized frequency estimation in federated settings, significantly improving accuracy and efficiency over existing methods.

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

PARD: a novel permutation-invariant autoregressive diffusion model for efficient and high-quality graph generation, achieving state-of-the-art results.

Vision Transformers achieve surprisingly high accuracy by transferring only pre-training attention maps, challenging the conventional belief that feature learning is crucial.

Researchers achieve full parameter identifiability in partially observed linear causal models using novel graphical conditions and a likelihood-based estimation method, addressing previous limitations…

Public data boosts private AI accuracy even with extreme distribution shifts, improving private model training by up to 67% in three tasks. This is due to shared low-dimensional representations betwe…

Omnigrasp: A novel RL-based method enables simulated humanoids to grasp diverse objects and precisely follow complex trajectories, advancing realistic human-object interaction in virtual environments.

OASIS, a novel data-centric approach, shapes offline data distributions toward safer, higher-reward policies using a conditional diffusion model, outperforming existing offline safe RL methods.

LLM watermarking faces inherent trade-offs; this paper reveals simple attacks exploiting common design choices, proposing guidelines and defenses for more secure systems.

In multi-agent imitation learning, achieving regret equivalence is harder than value equivalence; this paper introduces novel algorithms that efficiently minimize the regret gap under various assumpti…

Model-Based Diffusion (MBD) uses diffusion processes and model information for data-free trajectory optimization, outperforming existing methods on complex tasks.

MiSO: a novel closed-loop brain stimulation framework optimizes stimulation parameters to achieve desired neural population activity states, overcoming limitations of current methods by merging data a…

MGF: Mixed Gaussian Flow enhances trajectory prediction by using a mixed Gaussian prior, achieving state-of-the-art diversity and alignment accuracy.

Humans as landmarks: A novel zero-shot monocular metric depth estimation method leverages generative models and human mesh recovery to transfer metric scale information, achieving superior generalizat…

Researchers developed a new, differentiable score function for learning causal relationships from data that reliably recovers the simplest causal model, even with complex data.

MAmmoTH2: Harvesting 10M web instructions for enhanced LLM reasoning!

LLM dataset inference reliably detects if a dataset was used in training, overcoming limitations of existing membership inference attacks.

LipFD: a novel method leverages audio-visual inconsistencies to accurately spot lip-syncing deepfakes, outperforming existing methods and introducing a high-quality dataset for future research.