EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic . . . A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like “clever but clueless interns” in novel environments This severely limits their practical utility To systematically measure and drive progress on this challenge, we first introduce the Jericho Test-Time Learning (J-TTL) benchmark J-TTL is a new evaluation
DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining The idea to optimize the pre-training data sounds clever, particularly their use of a proxy model trained with DRO This offers a novel approach to learning the optimal data distribution without the costly estimation of the large pre-trained model’s task performance under different pretraining data settings
Hyperbolic Attention Networks - OpenReview We propose to incorporate inductive biases and operations coming from hyperbolic geometry to improve the attention mechanism of the neural networks
The Clever Hans Mirage: A Comprehensive Survey on Spurious. . . This survey on spurious correlations uses the Clever Hans metaphor to motivate the problem, formalizes a group-based setup g= (y,a) with core metrics (worst-group, average-group, bias-conflicting), and explains why models latch onto shortcuts (simplicity bias, training dynamics)
Building Normalizing Flows with Stochastic Interpolants A method is proposed to construct normalizing flows based on stochastic interpolants, yielding an efficient training algorithm compared to equivalent ODE methods, and providing a theoretical
Evaluating the Robustness of Neural Networks: An Extreme Value. . . Our analysis yields a novel robustness metric called CLEVER, which is short for Cross Lipschitz Extreme Value for nEtwork Robustness The proposed CLEVER score is attack-agnostic and is computationally feasible for large neural networks