1 citations · 1 across the 11 of their papers we have counts for
11 papers
Addressable Memory for Video World Models
Xindi Wu, Sven Elflein, James Lucas +5
We study visual persistence in interactive video world models. These models rely on a Key-Value (KV) cache as a growing visual memory to carry forward previously generated frames.…
Variance Reduction for Expectations with Diffusion Teachers
Jesse Bettencourt, Xindi Wu, Matan Atzmon +2
Pretrained diffusion models serve as frozen teachers feeding downstream pipelines such as text-to-3D, single-step distillation, and data attribution. The teacher gradients these pi…
Reinforced Fast Weights with Next-Sequence Prediction
Hee Seung Hwang, Xindi Wu, Sanghyuk Chun +1
Fast weight architectures offer a promising alternative to attention-based transformers for long-context modeling by maintaining constant memory overhead regardless of context leng…
Motion Attribution for Video Generation
Xindi Wu, Despoina Paschalidou, Jun Gao +5
Despite the rapid progress of video generation models, the role of data in influencing motion is poorly understood. We present Motive (MOTIon attribution for Video gEneration), a m…
Beyond Objects: Contextual Synthetic Data Generation for Fine-Grained Classification
William Yang, Xindi Wu, Zhiwei Deng +2
Text-to-image (T2I) models are increasingly used for synthetic dataset generation, but generating effective synthetic training data for classification remains challenging. Fine-tun…
Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting
Asher J. Hancock, Xindi Wu, Lihan Zha +2
Fine-tuning vision-language models (VLMs) on robot teleoperation data to create vision-language-action (VLA) models is a promising paradigm for training generalist policies, but it…
Explain Before You Answer: A Survey on Compositional Visual Reasoning
Fucai Ke, Joy Hsu, Zhixi Cai +10
Compositional visual reasoning has emerged as a key research frontier in multimodal AI, aiming to endow machines with the human-like ability to decompose visual scenes, ground inte…
DD-Ranking: Rethinking the Evaluation of Dataset Distillation
Zekai Li, Xinhao Zhong, Samir Khaki +49
In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance co…
Visual Compositional Tuning
Xindi Wu, Hee Seung Hwang, Polina Kirichenko +2
Visual instruction tuning (VIT) datasets have grown rapidly in scale, yet the informativeness of individual training samples has largely been overlooked. Recent dataset selection m…
ICONS: Influence Consensus for Vision-Language Data Selection
Xindi Wu, Mengzhou Xia, Rulin Shao +3
Training vision-language models via instruction tuning relies on large data mixtures spanning diverse tasks and domains, yet these mixtures frequently include redundant information…
ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty
Xindi Wu, Dingli Yu, Yangsibo Huang +2
Compositionality is a critical capability in Text-to-Image (T2I) models, as it reflects their ability to understand and combine multiple concepts from text descriptions. Existing e…