Archives AI News

Your Pre-trained LLM is Secretly an Unsupervised Confidence Calibrator

arXiv:2505.16690v3 Announce Type: replace Abstract: Post-training of large language models is essential for adapting pre-trained language models (PLMs) to align with human preferences and downstream tasks. While PLMs typically exhibit well-calibrated confidence, post-trained language models (PoLMs) often suffer from over-confidence,…

Evaluating multiple models using labeled and unlabeled data

arXiv:2501.11866v3 Announce Type: replace Abstract: It remains difficult to evaluate machine learning classifiers in the absence of a large, labeled dataset. While labeled data can be prohibitively expensive or impossible to obtain, unlabeled data is plentiful. Here, we introduce Semi-Supervised…

A Generalized Information Bottleneck Theory of Deep Learning

arXiv:2509.26327v2 Announce Type: replace Abstract: The Information Bottleneck (IB) principle offers a compelling theoretical framework to understand how neural networks (NNs) learn. However, its practical utility has been constrained by unresolved theoretical ambiguities and significant challenges in accurate estimation. In…

Offline Fictitious Self-Play for Competitive Games

arXiv:2403.00841v2 Announce Type: replace-cross Abstract: Offline Reinforcement Learning (RL) enables policy improvement from fixed datasets without online interactions, making it highly suitable for real-world applications lacking efficient simulators. Despite its success in the single-agent setting, offline multi-agent RL remains a…

LayerSync: Self-aligning Intermediate Layers

arXiv:2510.12581v1 Announce Type: cross Abstract: We propose LayerSync, a domain-agnostic approach for improving the generation quality and the training efficiency of diffusion models. Prior studies have highlighted the connection between the quality of generation and the representations learned by diffusion…

Y-shaped Generative Flows

arXiv:2510.11955v1 Announce Type: new Abstract: Modern continuous-time generative models often induce V-shaped transport: each sample travels independently along nearly straight trajectories from prior to data, overlooking shared structure. We introduce Y-shaped generative flows, which move probability mass together along shared…