arXiv:2605.10973v1 Announce Type: new
Abstract: Supervised fine-tuning (SFT) improves in-domain performance but can degrade out-of-domain (OOD) generalization. Prior work suggests that this degradation is related to changes in dominant singular subspaces of pretrained weight matrices. However, dire...
A Cascaded Generative Approach for e-Commerce Recommendations
arXiv:2605.11118v1 Announce Type: new
Abstract: Personalized storefronts in large e-commerce marketplaces are often assembled from many independent components: static themes per page section ("placement"), retrieval systems to fetch eligible products per placement, and pointwise rankers to order co...
RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking
arXiv:2605.11151v1 Announce Type: new
Abstract: Offline-to-online reinforcement learning (RL) improves sample efficiency by leveraging pre-collected datasets prior to online interaction. A key challenge, however, is learning an accurate critic in large state--action spaces with limited dataset cove...
Distributional Reinforcement Learning via the Cram\'er Distance
arXiv:2605.08104v1 Announce Type: new
Abstract: This paper explores the application of the Soft Actor-Critic (SAC) algorithm within a Distributional Reinforcement Learning setting and introduces an implementation of such algorithm named Cram\'er-based Distributional Soft Actor-Critic (C-DSAC). The ...
Where Reliability Lives in Vision-Language Models: A Mechanistic Study of Attention, Hidden States, and Causal Circuits
arXiv:2605.08200v1 Announce Type: new
Abstract: A pervasive intuition holds that vision-language models (VLMs) are most trustworthy when their attention maps look sharp: concentrated attention on the queried region should imply a confident, calibrated answer. We test this Attention-Confidence Assum...
arXiv:2605.08360v1 Announce Type: new
Abstract: Modern AI is opening the door to collective decision-making in which participants express their views as free-form text rather than voting on a fixed set of candidates. A natural idea is to embed these opinions in a vector space so that the substantia...
Researchers Worldwide Compete to Shape the Future of AI in Organizations
More than 200 academic teams submitted proposals to the AI for Organizations Grand Challenge, exploring how artificial intelligence will transform teamwork and collaboration.
SocialReasoning-Bench: Measuring whether AI agents act in users’ best interests
Using SocialReasoning Bench, we observed a stable pattern across models—agents execute competently, but fail to consistently improve the user’s position, even with explicit instructions to optimize for user interest.
The post SocialReasoning-Bench: Measuring whether AI agents act in users’ best inte...
On the Role of Strain and Vorticity in Numerical Integration Error for Flow Matching
arXiv:2605.06680v1 Announce Type: new
Abstract: Flow matching generates data by integrating a learned velocity field, where the number of integration steps (NFE) directly determines inference cost. We analyze which properties of the velocity field govern integration error by decomposing the velocit...
arXiv:2605.05365v1 Announce Type: new
Abstract: We present ZAYA1-8B, a reasoning-focused mixture-of-experts (MoE) model with 700M active and 8B total parameters, built on Zyphra's MoE++ architecture. ZAYA1-8B's core pretraining, midtraining, and supervised fine-tuning (SFT) were performed on a full...
arXiv:2605.05209v1 Announce Type: new
Abstract: Neural networks that land in flat regions of the loss landscape tend to generalise better than those in sharp regions. Sharpness-Aware Minimisation exploits this to improve generalisation. But function-preserving reparameterisation can inflate the Hes...
Horizon-Constrained Rashomon Sets for Chaotic Forecasting
arXiv:2605.05218v1 Announce Type: new
Abstract: Predictive multiplicity and chaotic dynamics represent two fundamental challenges in machine learning that have evolved independently despite their conceptual connections. We bridge this gap by introducing horizon-constrained Rashomon sets, a theoreti...
Large-Scale High-Quality 3D Gaussian Head Reconstruction from Multi-View Captures
We propose HeadsUp, a scalable feed-forward method for reconstructing high-quality 3D Gaussian heads from large-scale multi-camera setups. Our method employs an efficient encoder-decoder architecture that compresses input views into a compact latent representation. This latent representation is then...
Apple Workshop on Privacy-Preserving Machine Learning & AI 2026
At Apple, we believe privacy is a fundamental human right. As AI capabilities increase and become more integrated into people’s daily lives, advancing research in privacy-preserving techniques is increasingly important to ensure privacy is protected while users enjoy innovative AI experiences.
Appl...
A Self-Attentive Meta-Optimizer with Group-Adaptive Learning Rates and Weight Decay
arXiv:2605.04055v1 Announce Type: new
Abstract: Adaptive optimizers like AdamW apply uniform hyperparameters across all parameter groups, ignoring heterogeneous optimization dynamics across layers and modules. We address this limitation by proposing MetaAdamW - a new optimizer that integrates a sel...
arXiv:2605.04100v1 Announce Type: new
Abstract: Off-policy temporal-difference (TD) learning with function approximation faces a structural tradeoff among stability, projection geometry, and variance control. Emphatic TD (ETD) improves the off-policy projection geometry through follow-on emphasis, ...
ANDRE: An Attention-based Neuro-symbolic Differentiable Rule Extractor
arXiv:2605.04193v1 Announce Type: new
Abstract: Inductive Logic Programming (ILP) aims to learn interpretable first-order rules from data, but existing symbolic and neuro-symbolic approaches struggle to scale to noisy and probabilistic settings. Classical ILP relies on discrete combinatorial rule s...
What Matters in Practical Learned Image Compression
One of the major differentiators unlocked by learned codecs relative to their hard-coded traditional counterparts is their ability to be optimized directly to appeal to the human visual system. Despite this potential, a perceptual yet practical image codec is yet to be proposed. In this work, we aim...
Text-Conditional JEPA for Learning Semantically Rich Visual Representations
Image-based Joint-Embedding Predictive Architecture (I-JEPA) offers a promising approach to visual self-supervised learning through masked feature prediction. However with the inherent visual uncertainty at masked positions, feature prediction remains challenging and may fail to learn semantic repre...