Publications (11)
Methods and strategies for improving the novel view synthesis quality of neural radiation field
Shun Fang, Ming Cui, Xing Feng +1
Neural Radiation Field (NeRF) technology can learn a 3D implicit model of a scene from 2D images and synthesize realistic novel view images. This technology has received widespread…
Exploration and Improvement of Nerf-based 3D Scene Editing Techniques
Shun Fang, Ming Cui, Xing Feng +1
NeRF's high-quality scene synthesis capability was quickly accepted by scholars in the years after it was proposed, and significant progress has been made in 3D scene representatio…
DUMA: a Dual-Mind Conversational Agent with Fast and Slow Thinking
Xiaoyu Tian, Liangyu Chen, Na Liu +4
Inspired by the dual-process theory of human cognition, we introduce DUMA, a novel conversational agent framework that embodies a dual-mind mechanism through the utilization of two…
DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training
Haisen Luo, Yiwei Liu, Haoning Wang +13
Enabling large language models to achieve stable self-improvement without external expert supervision remains a central challenge in complex reasoning tasks. Existing self-distilla…
Weakly supervised alignment and registration of MR-CT for cervical cancer radiotherapy
Jjahao Zhang, Yin Gu, Deyu Sun +5
Cervical cancer is one of the leading causes of death in women, and brachytherapy is currently the primary treatment method. However, it is important to precisely define the extent…
From LLM to Conversational Agent: A Memory Enhanced Architecture with Fine-Tuning of Large Language Models
Na Liu, Liangyu Chen, Xiaoyu Tian +3
This paper introduces RAISE (Reasoning and Acting through Scratchpad and Examples), an advanced architecture enhancing the integration of Large Language Models (LLMs) like GPT-4 in…
A real-time rendering method for high albedo anisotropic materials with multiple scattering
Shun Fang, Xing Feng, Ming Cui
We propose a neural network-based real-time volume rendering method for realistic and efficient rendering of volumetric media. The traditional volume rendering method uses path tra…
An efficient fully decoupled finite element method with second-order accuracy for the micropolar Rayleigh-Benard convection system
Ming Cui, Akang Hou, Xiaoyu Dong
The micropolar Rayleigh-B{é}nard convection system, which consists of Navier-Stokes equations, the angular momentum equation, and the heat equation, is a strongly nonlinear, coupl…
Sequential symmetric interior penalty discontinuous Galerkin method for fully coupled quasi-static thermo-poroelasticity problems
Fan Chen, Ming Cui, Chenguang Zhou
In this paper, we investigate a sequentially decoupled numerical method for solving the fully coupled quasi-static thermo-poroelasticity problems with nonlinear convective transpor…
ASTRA: Automated Synthesis of agentic Trajectories and Reinforcement Arenas
Xiaoyu Tian, Haotian Wang, Shuaiting Chen +12
Large language models (LLMs) are increasingly used as tool-augmented agents for multi-step decision making, yet training robust tool-using agents remains challenging. Existing meth…
Dial-insight: Fine-tuning Large Language Models with High-Quality Domain-Specific Data Preventing Capability Collapse
Jianwei Sun, Chaoyang Mei, Linlin Wei +4
The efficacy of large language models (LLMs) is heavily dependent on the quality of the underlying data, particularly within specialized domains. A common challenge when fine-tunin…