11 accepted papers
Recent advancements in Large Multimodal Models (LMMs) have significantly improved multimodal understanding and generation. However, these models still struggle to generate tightly interleaved image-text outputs, primarily due to the limited scale, quality and instructional richness of current traini…
A fundamental bottleneck in human-AI collaboration is the "intention expression gap", the difficulty for humans to effectively convey complex, high-dimensional thoughts to AI. This challenge often traps users in inefficient trial-and-error loops and is exacerbated by the diverse expertise levels of …
Online continual learning (OCL) aims at learning a non-stationary data stream in a way of reading each data sample only once, and hence suffers from the trade-off of catastrophic forgetting and insufficient learning. In this work, we firstly analytically establish relationship between loss functions
In spinal surgery, ensuring surgical precision and safety is paramount. Traditionally, surgeons have relied on their experience to determine when to cease milling as the cutter approaches the spinal cord; However, improper technique during this process can lead to complications, such as vertebral pl
Distillation strategies are currently the primary approaches for mitigating forgetting in class incremental learning (CIL). Existing methods generally inherit previous knowledge from a single teacher. However teachers with different mechanisms are talented at different tasks and inheriting diverse k…
Contrastive learning is a powerful paradigm for representation learning with prominent success in computer vision and NLP, but how to extend its success to high-dimensional tensors remains a challenge. This is because tensor data often exhibit high-order mode-interactions that are hard to profile an…
Large Vision-Language Models (VLMs) have already been applied to the understanding of Graphical User Interfaces (GUIs) and have achieved notable results. However, existing VLMs often overly rely on internal text-based knowledge while neglecting visual inputs. This imbalance may lead models to produc…
To optimize the cutting depth in robotic-assisted laminectomy, we present a real-time method to adjust the preoperatively planned feed rate in the depth direction of the robot cutting trajectory. Not only the linearity between the harmonic amplitude of the milling acoustic signal and the cutting dep…
Surgical navigation is challenging on complicated multi-branch structures such as intrarenal collecting systems or bronchi. The objective of this work is to help surgeons quickly establish the corresponding relationship between intraoperative endoscopic images and preoperative CT data. An endoscopic…
Bone milling is widely used in many hard tissue surgical procedures, and the main concern is the mechanical damage to some important tissues induced by the high-speed rotating tool. In this study, the behavior of the human auditory system is analyzed. Following this, a commercially available microph…
One of the major challenge in spine surgery is successfully cutting the desired bony structure while avoiding injury to the vital anatomy such as the spinal cord. This paper presents a vibration signal processing method to discriminate different types of tissue in robot-assisted spine surgery. Durin…