7 accepted papers
Sparse Autoencoders (SAEs) are widely used to interpret large language models by decomposing activations into sparse, human-understandable features, but scaling to large dictionaries exposes fundamental challenges. Systematic studies reveal pervasive feature splitting that fragments coherent concept…
Continuum and snake-like hyper-redundant robots perform circular bending motions, enable them to navigate obstacles and operate in confined spaces, making them ideal for applications such as industrial inspection and minimally invasive surgery. However, their performance often diminishes at the dist
Aligning Large Language Models (LLMs) with human values has attracted increasing attention since it provides clarity, transparency, and the ability to adapt to evolving scenarios. In this paper, we introduce a Controlled Value Vector Activation (ConVA) method that directly aligns the internal values…
Click-through rate (CTR) prediction is a vital task in industrial recommendation systems. Most existing methods focus on the network architecture design of the CTR model for better accuracy and suffer from the data sparsity problem. Especially in industrial recommendation systems, the widely applied…
With the growing popularity of general-purpose Large Language Models (LLMs), comes a need for more global explanations of model behaviors. Concept-based explanations arise as a promising avenue for explaining high-level patterns learned by LLMs. Yet their evaluation poses unique challenges, especial…
Ultrasound (US) imaging has been widely utilized for guiding puncture procedures in spinal interventions. However, conventional methods that confine puncture within the US plane may not adequately meet the demands of endoscopic spinal surgery (ESS), which requires extensive spatial orientation adjus
Robot-assisted ultrasound scanning acquires spatial-property labeled images under high consistency, holding the potential to reconstruct the volume with more intuitive images for examination and surgery guidance. Promisingly, the vertebral level identification, as a time-consuming procedure before s