A deep dictionary network-based foundation model for ultra-low-dose CT denoising
为解决超低剂量CT图像噪声问题，提出基于深度字典网络的统一多器官去噪基础模型，通过预训练和微调实现跨区域去噪。
0 citationsRead paper
Selected work
Representative Papers
The Geometry of Persona: Disentangling Personality from Reasoning in Large Language Models
Personalized large language models face a fundamental stability–plasticity trade-off: existing alignment methods (e.g., supervised fine-tuning) impose an “alignment tax,” degrading general reasoning capabilities. To address this, we propose Soul Engine—a novel framework that geometrically models personality as a linear, orthogonal subspace to the reasoning subspace within a frozen backbone model, thereby decoupling personality from core reasoning ability. Our approach employs a dual-head architecture, a dynamically context-sampled benchmark dataset (SoulBench), and vector-arithmetic-based personality modulation—enabling zero-shot personality injection and deterministic behavioral control. Experiments demonstrate high-fidelity personality modeling (MSE = 0.011) and manifold-level orthogonality and continuity, verified via T-SNE. Crucially, Soul Engine achieves controllable, high-fidelity personality customization without backbone fine-tuning, preserving full reasoning capability—eliminating the alignment tax while enabling lossless, interpretable personalization.
0 citationsRead paper
Recent publications
Latest Papers
A deep dictionary network-based foundation model for ultra-low-dose CT denoising
为解决超低剂量CT图像噪声问题，提出基于深度字典网络的统一多器官去噪基础模型，通过预训练和微调实现跨区域去噪。
0 citationsRead paper
The Geometry of Persona: Disentangling Personality from Reasoning in Large Language Models
Personalized large language models face a fundamental stability–plasticity trade-off: existing alignment methods (e.g., supervised fine-tuning) impose an “alignment tax,” degrading general reasoning capabilities. To address this, we propose Soul Engine—a novel framework that geometrically models personality as a linear, orthogonal subspace to the reasoning subspace within a frozen backbone model, thereby decoupling personality from core reasoning ability. Our approach employs a dual-head architecture, a dynamically context-sampled benchmark dataset (SoulBench), and vector-arithmetic-based personality modulation—enabling zero-shot personality injection and deterministic behavioral control. Experiments demonstrate high-fidelity personality modeling (MSE = 0.011) and manifold-level orthogonality and continuity, verified via T-SNE. Crucially, Soul Engine achieves controllable, high-fidelity personality customization without backbone fine-tuning, preserving full reasoning capability—eliminating the alignment tax while enabling lossless, interpretable personalization.
0 citationsRead paper