AI Theory

PamaCF, a novel personalized adversarial collaborative filtering technique, significantly improves recommendation robustness and accuracy against poisoning attacks by dynamically adjusting perturbatio…

Current backdoor defenses, while effective at reducing attack success rates, are vulnerable to rapid re-learning. This work unveils this superficial safety, proposes a novel attack, and introduces a p…

A revolutionary ultrafast phylogenetic method outperforms protein language models in variant effect prediction by efficiently estimating amino acid substitution rates from massive datasets.

A unified neural divide-and-conquer framework (UDC) achieves superior performance on large-scale combinatorial optimization problems by employing a novel Divide-Conquer-Reunion training method and a h…

Researchers developed Subsampled Smooth Calibration Error (SSCE), a new truthful calibration measure for sequential prediction, solving the problem of existing measures being easily gamed.

This paper presents a novel, truthful, and privacy-preserving mechanism for high-dimensional sparse linear regression, incentivizing data contribution while safeguarding individual privacy.

This paper optimizes randomized smoothing, a crucial certified defense against adversarial attacks, by introducing novel statistical methods that drastically reduce the computational cost, leading to …

Trap-MID: Outsmarting model inversion attacks with cleverly placed ’trapdoors'!

This paper introduces a novel theoretical framework for robust machine learning under distribution shifts, offering learning rules and guarantees, highlighting the game-theoretic viewpoint of distribu…

Challenging common assumptions, researchers prove that flatter adversarial examples don’t guarantee better transferability and introduce TPA, a theoretically-grounded attack creating more transferable…

Supervised learning’s sample complexity is compact: a hypothesis class is learnable if and only if all its finite projections are learnable, simplifying complexity analysis.

Generative models can outperform their human trainers: A groundbreaking study shows how autoregressive transformers, trained on chess game data, can achieve higher game ratings than any of the human …

R2ET: training for robust ranking explanations by an effective regularizer.

Boosting hippocampal spatial resolution surprisingly shrinks its contextual memory capacity, revealing a crucial trade-off between precision and context storage.

Researchers achieve a balance between accuracy and efficiency in multiclass classification by introducing partially consistent surrogate losses and novel methods.

This paper examines the trade-offs between two popular diagonal Fisher Information Matrix (FIM) estimators in neural networks, deriving variance bounds and highlighting the importance of considering e…

Trace: Automating AI workflow design with LLMs.

DNNs learn interactions in two phases: initially removing complex interactions, then gradually learning higher-order ones, leading to overfitting.

VFair achieves harmless Rawlsian fairness in regression tasks without relying on sensitive demographic data by minimizing the variance of training losses.

This paper presents a novel framework for estimating bounds on policy effects under unobserved confounding, offering tighter bounds and robust estimators for higher-dimensional data.