Chu Wang

Yuanhao Cai, Kunpeng Li, Menglin Jia et al.

Recent advances in text-to-video (T2V) generation have achieved good visual quality, yet synthesizing videos that faithfully follow physical laws remains an open challenge. Existing methods mainly based on graphics or prompt extension struggle to generalize beyond simple simulated environments or learn implicit physical reasoning. The scarcity of training data with rich physics interactions and phenomena is also a problem. In this paper, we first introduce a Physics-Augmented video data construction Pipeline, PhyAugPipe, that leverages a vision-language model (VLM) with chain-of-thought reasoning to collect a large-scale training dataset, PhyVidGen-135K. Then we formulate a principled Physics-aware Groupwise Direct Preference Optimization, PhyGDPO, framework that uses real-world video as winning case to guarantee correct physics learning and builds upon the groupwise Plackett-Luce probabilistic model to capture holistic preferences beyond pairwise comparisons. In PhyGDPO, we design a Physics-Guided Rewarding (PGR) scheme that leverages VLM-based physical rewards to direct the optimization to focus on challenging physics cases. In addition, we propose a LoRA-Switch Reference (LoRA-SR) scheme that avoids full-model duplication as reference for efficient DPO training. Experiments show that our method significantly outperforms state-of-the-art open-source methods on PhyGenBench and VideoPhy2. Please check our project page at https://caiyuanhao1998.github.io/project/PhyGDPO for more video results. Our code, models, and data will be released at https://github.com/caiyuanhao1998/Open-PhyGDPO

Chu Wang, Marcello Pelillo, Kaleem Siddiqi

View based strategies for 3D object recognition have proven to be very successful. The state-of-the-art methods now achieve over 90% correct category level recognition performance on appearance images. We improve upon these methods by introducing a view clustering and pooling layer based on dominant sets. The key idea is to pool information from views which are similar and thus belong to the same cluster. The pooled feature vectors are then fed as inputs to the same layer, in a recurrent fashion. This recurrent clustering and pooling module, when inserted in an off-the-shelf pretrained CNN, boosts performance for multi-view 3D object recognition, achieving a new state of the art test set recognition accuracy of 93.8% on the ModelNet 40 database. We also explore a fast approximate learning strategy for our cluster-pooling CNN, which, while sacrificing end-to-end learning, greatly improves its training efficiency with only a slight reduction of recognition accuracy to 93.3%. Our implementation is available at https://github.com/fate3439/dscnn.

Yuhang Wu, Yingfei Wang, Chu Wang et al.

Pre-trained large language models (LLM) have emerged as a powerful tool for simulating various scenarios and generating output given specific instructions and multimodal input. In this work, we analyze the specific use of LLM to enhance a classical supervised machine learning method for classification problems. We propose a few approaches to integrate LLM into a classical machine learning estimator to further enhance the prediction performance. We examine the performance of the proposed approaches through both standard supervised learning binary classification tasks, and a transfer learning task where the test data observe distribution changes compared to the training data. Numerical experiments using four publicly available datasets are conducted and suggest that using LLM to enhance classical machine learning estimators can provide significant improvement on prediction performance.

Chu Wang, Lei Tang, Yang Lu et al.

ProductNet is a collection of high-quality product datasets for better product understanding. Motivated by ImageNet, ProductNet aims at supporting product representation learning by curating product datasets of high quality with properly chosen taxonomy. In this paper, the two goals of building high-quality product datasets and learning product representation support each other in an iterative fashion: the product embedding is obtained via a multi-modal deep neural network (master model) designed to leverage product image and catalog information; and in return, the embedding is utilized via active learning (local model) to vastly accelerate the annotation process. For the labeled data, the proposed master model yields high categorization accuracy (94.7% top-1 accuracy for 1240 classes), which can be used as search indices, partition keys, and input features for machine learning models. The product embedding, as well as the fined-tuned master model for a specific business task, can also be used for various transfer learning tasks.

Chu Wang, Lei Tang, Shujun Bian et al.

For a product of interest, we propose a search method to surface a set of reference products. The reference products can be used as candidates to support downstream modeling tasks and business applications. The search method consists of product representation learning and fingerprint-type vector searching. The product catalog information is transformed into a high-quality embedding of low dimensions via a novel attention auto-encoder neural network, and the embedding is further coupled with a binary encoding vector for fast retrieval. We conduct extensive experiments to evaluate the proposed method, and compare it with peer services to demonstrate its advantage in terms of search return rate and precision.