7 papers
Logic-Regularized Verifier Elicits Reasoning from LLMs
Xinyu Wang, Changzhi Sun, Lian Cheng +4
Verifiers are crucial components for enhancing modern LLMs' reasoning capability. Typicalverifiers require resource-intensive superviseddataset construction, which is costly and fa…
Beyond Heuristics: A Decision-Theoretic Framework for Agent Memory Management
Changzhi Sun, Xiangyu Chen, Jixiang Luo +2
External memory is a key component of modern large language model (LLM) systems, enabling long-term interaction and personalization. Despite its importance, memory management is st…
Gestura: A LVLM-Powered System Bridging Motion and Semantics for Real-Time Free-Form Gesture Understanding
Zhuoming Li, Aitong Liu, Mengxi Jia +5
Free-form gesture understanding is highly appealing for human-computer interaction, as it liberates users from the constraints of predefined gesture categories. However, the sole e…
EV-NVC: Efficient Variable bitrate Neural Video Compression
Yongcun Hu, Yingzhen Zhai, Jixiang Luo +4
Training neural video codec (NVC) with variable rate is a highly challenging task due to its complex training strategies and model structure. In this paper, we train an efficient v…
SmartFreeEdit: Mask-Free Spatial-Aware Image Editing with Complex Instruction Understanding
Qianqian Sun, Jixiang Luo, Dell Zhang +1
Recent advancements in image editing have utilized large-scale multimodal models to enable intuitive, natural instruction-driven interactions. However, conventional methods still f…
Infinite Video Understanding
Dell Zhang, Xiangyu Chen, Jixiang Luo +6
The rapid advancements in Large Language Models (LLMs) and their multimodal extensions (MLLMs) have ushered in remarkable progress in video understanding. However, a fundamental ch…