…Xiyu Ren , , Yiming Du , , , , , , , , , , , Abstract A new benchmark evaluates memory capabilities in vision-language models through multi-session conversations, revealing limitations of both long-context and memory-augmented approaches. AI-generated summary…
…We introduce JLT, a 130M latent diffusion Transformer over frozen FLUX.2 VAE codes, and compare clean-latent prediction with a matched velocity-prediction DiT under the same representation, backbone, and training…
…Inference-Time Mixed-Precision Quantization for Video Diffusion Models (2026) Streaming Autoregressive Video Generation via Diagonal Distillation (2026) PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference (2026…
…Continual Learning from Experience and Skills in Multimodal Agents (2026) Training-Free Test-Time Contrastive Learning for Large Language Models (2026) OSExpert: Computer-Use Agents Learning Professional Skills via Exploration (2026) WebXSkill…
…AI-generated summary Despite rapid advancements, current text-to-image (T2I) models predominantly rely on a single-step generation paradigm, which struggles with complex semantics and faces diminishing returns from parameter scaling…
To show you the most relevant results, we’ve omitted some entries very similar to those already shown. Repeat the search with the omitted results included.