Papers arxiv:2605.07079 Learning Visual Feature-Based World Models via Residual Latent Action Published on May 8 Submitted by Xinyu Zhang on May 11 Authors: , Zhengtong Xu , , , , Abstract Visual world models…
…Topologically Aligned Encoding of Vision-Transformer Feature Spaces (2026) MacTok: Robust Continuous Tokenization for Image Generation (2026) Coevolving Representations in Joint Image-Feature Diffusion (2026) Repurposing Geometric Foundation Models for Multi-view…
…The exact boundary is only one of many feature sets that beat all features. We then develop what these facts imply for prediction-aligned feature selection and for tabular models that learn…
…Abstract Linear probes for deception detection in large language models fail under distributional shifts despite high performance on clean data, revealing that deception is encoded through distributed sub-threshold features rather than…
Papers arxiv:2606.04048 Unlocking Feature Learning in Gated Delta Networks at Scale Published on Jun 2 Submitted by Yifeng Liu on Jun 4 University of California, Los Angeles Authors: Yifeng Liu…
…The recovered features are interpretable, spanning phonemes , laughter , accent prompts and speaker gender . Steering through the SAE latent space shows these features are causal rather than merely descriptive: targeted interventions raise laughter…
…WildRelight features a diverse collection of high-resolution outdoor scenes, captured under strictly aligned, temporally varying natural illuminations, each paired with a high-dynamic-range environment map. Using this data, we establish…
…Selim Kuzucu , , , , , Abstract PARCEL is a vision-language model architecture that dynamically partitions feature extraction tasks to improve efficiency and performance across different visual-token budgets. Generated by Qwen/Qwen2.5-Coder…
…Unlike standard vision-essential benchmarks that evaluate a single input form, SeePhys Pro features four semantically aligned variants for each problem with progressively increasing visual elements. Our evaluation shows that current frontier…
…By aligning intermediate diffusion features with clean-image representations from self-supervised vision encoders , existing methods improve convergence and generation quality. However, such alignment also introduces a non-trivial constraint: diffusion models…