11 accepted papers
Automatic medical report generation can greatly reduce the workload of doctors, but it is often unreliable for real-world deployment. Current methods can write formally fluent sentences but may be factually flawed, introducing serious medical errors known as clinical hallucinations, which make them
Video foundation models aim to integrate video understanding, generation, editing, and instruction following within a single framework, making them a central direction for next-generation multimodal systems. However, existing evaluation benchmarks remain fragmented and limited in scope, as they each
Inference-time alignment provides an efficient alternative for aligning LLMs with humans. However, these approaches still face challenges, such as limited scalability due to policy-specific value functions and latency during the inference phase. In this paper, we propose a novel approach, Diffusion-…
Anomaly detection (AD) is essential for industrial inspection, yet existing methods typically rely on "comparing" test images to normal references from a training set. However, variations in appearance and positioning often complicate the alignment of these references with the test image, limiting d…
Understanding neural network' (NN) generalizability remains a central question in deep learning research. The special phenomenon of grokking, where NNs abruptly generalize long after the training performance reaches near-perfect level, offers a unique window to investigate the underlying mechanisms…
Aligning with personalized preferences, which vary significantly across cultural, educational, and political differences, poses a significant challenge due to the computational costs and data demands of traditional alignment methods. In response, this paper presents Personalized Alignment at Decodin…
Aligning language models with human preferences presents significant challenges, particularly in achieving personalization without incurring excessive computational costs. Existing methods rely on reward signals and additional annotated data, limiting their scalability and adaptability to diverse hu…
While pre-trained Chinese language models have demonstrated impressive performance on a wide range of NLP tasks, the Chinese Spell Checking (CSC) task remains a challenge. Previous research has explored using information such as glyphs and phonetics to improve the ability of CSC models to distinguis…
Dialogue data in real scenarios tend to be sparsely available, rendering data-starved end-to-end dialogue systems trained inadequately. We discover that data utilization efficiency in low-resource scenarios can be enhanced by mining alignment information uncertain utterance and deterministic dialogu…
This paper presents the design and testing of a bio-inspired soft pneumatic robot that can achieve locomotion along the outside of a cylinder. The robot uses soft pneumatic actuators called FREEs (Fiber Reinforced Elastomeric Enclosure), which can have a wide range of deformation behavior upon press…
Soft robotic actuators are well suited for use in exoskeleton applications due to their innate compliance and low weight. We have developed a wearable soft robotic sleeve that uses fiber reinforced elastomeric enclosures (FREEs) to provide actuation and stiffness at the elbow for augmented lifting a…