MORPHOS is a novel autoregressive framework that generates dynamic 3D assets (like meshes and radiance fields) from videos by using a unified 4D representation to ensure temporal consistency and handl…

The paper proposes VRPO, a reinforcement learning-based optimization strategy that replaces static alignment losses in diffusion models, significantly improving both convergence and image fidelity.

This paper introduces Flex-Forcing, a framework for video generation that enables a model to operate under both bidirectional and autoregressive generation regimes, achieving better video quality and…

This paper proposes ReChannel, a method for dense prediction using a pretrained DiT model, which keeps the encoder but removes the decoder and adapts it with task LoRA. ReChannel maps each token to it…

This paper analyzes existing watermarking schemes for autoregressive image generators and demonstrates that they are vulnerable to various removal and forgery attacks, suggesting they are unreliable f…

This paper proposes Twins, a unified continuous token space for multimodal models using ViT and VAE features, and addresses optimization imbalance with a focal regression objective.

DigenRL is a disaggregated RL framework for diffusion-based generative LLMs that achieves 1.56-2.10x throughput improvements over state-of-the-art diffusion RL systems.

LongLive-RAG proposes a novel Retrieval-Augmented Generation (RAG) framework to stabilize and improve the quality of long-horizon video generation by treating the entire generated history as a searcha…

This paper introduces Align4D, a framework for generating coherent video-3D pairs using any-modal input, achieving state-of-the-art quality and consistency in X-to-4D generation.

This paper introduces MultiHashFormer, a framework for hash-based autoregression in language models, which allows for parameter efficiency while avoiding many-to-one collisions.

The paper introduces Drifting Preference Optimization (DrPO), an efficient online method for preference finetuning one-step text-to-image generators that avoids complex gradient calculations and model…

This paper proposes a semi-supervised framework, RepG, for conditional generative modeling using stochastic interpolation and low-dimensional latent representations.

The paper analyzes token reduction for efficient unified VLM training, finding that while task-specific acceleration saves computation, it destroys the mutual performance gains achieved through joint…

The paper introduces LL-Bench, a comprehensive benchmark for evaluating large-scale generative models on low-level vision tasks, and proposes LL-Score, an MLLM-based evaluator that better aligns quali…

The paper proposes Alignment-Guided Score Matching (AGSM), a lightweight, reward-free post-training method that integrates contrastive alignment guidance directly into the score-matching objective of…

The paper introduces DLM-SWAI, a training-free method that effectively steers diffusion language models (DLMs) toward desired textual styles or properties by biasing the token distribution at each den…

The paper introduces GPIC, a massive, permissively licensed, and safety-filtered image corpus of 28 trillion pixels, designed to serve as a stable and accessible benchmark for large-scale visual gener…

VISReg introduces a novel regularization technique that combines variance control with a Sliced-Wasserstein-based sketching objective to stabilize self-supervised learning, achieving state-of-the-art…