…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…
…Xuanyu Zhu , , Yang Shi , , , , Abstract DRoRAE enhances visual representation by fusing multi-layer features from pretrained vision encoders through adaptive routing and incremental correction, improving reconstruction and generation quality. AI-generated summary…
…Specifically, during training, we make the model acts as both the teacher and the student with different contexts, where the student is conditioned only on the text feature , while the teacher is…
…providing voxel features as vertex representations, guiding token prediction via cross-attention to voxel features, and serving as a structural scaffold that constrains generation around the input surface. Our hierarchical design enables…
…To support this paradigm, we curate and release GEM-4M , a comprehensive large-scale dataset featuring a mixture of grounding, reasoning, and planning data paired with high-quality depth supervision. Extensive experiments…
…By systematically decorrelating provability from spurious features and enforcing algorithmic alignment , we find that in sufficiently deep models with a bidirectional prefix mask , implicit reasoning approaches explicit CoT performance across graph topologies…
…We adopt the E4SRec framework by extending it with multimodal features and different item ID encoder backbones , including SASRec , BERT4Rec , and GRU4Rec . We further extend the LLM backbone option with LLaMa-2…
…Systematic Benchmarking and Improving Omnimodal Personalization Published on May 11 Submitted by Yeongtak on May 12 Seoul National University Authors: , , , , Abstract Omni-Persona introduces the first comprehensive benchmark for omnimodal personalization, featuring…
…Physics-Coherent Image-to-Video Generation through Feature and 3D Representation Alignment (2026) Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion (2026) Toward Physically Consistent Driving Video World Models under…