6 accepted papers
Surgical phase recognition is critical in computer-assisted surgery. Clinically, surgeons discriminate surgical phases through visuospatial analysis of instrument-tissue interactions. However, existing methods fail to adequately account for the critical role of the visual-neural mechanisms of the su
The emergence of Large Language Models (LLMs) has advanced the multilingual machine translation (MMT), yet the Curse of Multilinguality (CoM) remains a major challenge. Existing work in LLM-based MMT typically mitigates this issue via scaling up training and computation budget, which raises a critic…
Model quantization reduces the bit-width of weights and activations, improving memory efficiency and inference speed in diffusion models. However, achieving 4-bit quantization remains challenging. Existing methods, primarily based on integer quantization and post-training quantization fine-tuning, s…
SLAM is the key technique for localization and surrounding perception in indoor environments. However, the dynamic posture adjustments of wheel-legged robots cast new challenges that affect the accuracy of localization. Therefore, this letter presents the Wheel-Legged SLAM, a novel indoor SLAM metho
Reinforcement Learning (RL) problem with general utility is a powerful decision making framework that covers standard RL with cumulative cost, exploration problems, and demonstration learning. Existing works on RL with general utility do not consider the robustness under environmental perturbation,…