Fan Zhang

Xiaotong Li, Fan Zhang, Haiwen Diao et al.

Existing Multimodal Large Language Models (MLLMs) increasingly emphasize complex understanding of various visual elements, including multiple objects, text information, and spatial relations. Their development for comprehensive visual perception hinges on the availability of high-quality image-text datasets that offer diverse visual elements and throughout image descriptions. However, the scarcity of such hyper-detailed datasets currently hinders progress within the MLLM community. The bottleneck stems from the limited perceptual capabilities of current caption engines, which fall short in providing complete and accurate annotations. To facilitate the cutting-edge research of MLLMs on comprehensive vision perception, we thereby propose Perceptual Fusion, using a low-budget but highly effective caption engine for complete and accurate image descriptions. Specifically, Perceptual Fusion integrates diverse perception experts as image priors to provide explicit information on visual elements and adopts an efficient MLLM as a centric pivot to mimic advanced MLLMs' perception abilities. We carefully select 1M highly representative images from uncurated LAION dataset and generate dense descriptions using our engine, dubbed DenseFusion-1M. Extensive experiments validate that our engine outperforms its counterparts, where the resulting dataset significantly improves the perception and cognition abilities of existing MLLMs across diverse vision-language benchmarks, especially with high-resolution images as inputs. The dataset and code are publicly available at https://github.com/baaivision/DenseFusion.

Zihong He, Weizhe Lin, Hao Zheng et al.

With the rapid advancement of AI systems, their abilities to store, retrieve, and utilize information over the long term - referred to as long-term memory - have become increasingly significant. These capabilities are crucial for enhancing the performance of AI systems across a wide range of tasks. However, there is currently no comprehensive survey that systematically investigates AI's long-term memory capabilities, formulates a theoretical framework, and inspires the development of next-generation AI long-term memory systems. This paper begins by introducing the mechanisms of human long-term memory, then explores AI long-term memory mechanisms, establishing a mapping between the two. Based on the mapping relationships identified, we extend the current cognitive architectures and propose the Cognitive Architecture of Self-Adaptive Long-term Memory (SALM). SALM provides a theoretical framework for the practice of AI long-term memory and holds potential for guiding the creation of next-generation long-term memory driven AI systems. Finally, we delve into the future directions and application prospects of AI long-term memory.

Hammad Ud Din Ahmed, Usama Ijaz Bajwa, Fan Zhang et al.

Facts are important in decision making in every situation, which is why it is important to catch deceptive information before they are accepted as facts. Deception detection in videos has gained traction in recent times for its various real-life application. In our approach, we extract facial action units using the facial action coding system which we use as parameters for training a deep learning model. We specifically use long short-term memory (LSTM) which we trained using the real-life trial dataset and it provided one of the best facial only approaches to deception detection. We also tested cross-dataset validation using the Real-life trial dataset, the Silesian Deception Dataset, and the Bag-of-lies Deception Dataset which has not yet been attempted by anyone else for a deception detection system. We tested and compared all datasets amongst each other individually and collectively using the same deep learning training model. The results show that adding different datasets for training worsen the accuracy of the model. One of the primary reasons is that the nature of these datasets vastly differs from one another.