13 accepted papers
Concept erasure is extensively utilized in image generation to prevent text-to-image models from generating undesired content. Existing methods can effectively erase narrow concepts that are specific and concrete, such as distinct intellectual properties (e.g. Pikachu) or recognizable characters (e.
Reinforcement learning (RL) has emerged as a central paradigm for training large language models (LLMs) in reasoning tasks. Yet recent studies question RL’s ability to incentivize reasoning capacity beyond the base model. This raises a key challenge: how can RL be adapted to solve harder reasoning p…
Vision-Language-Action (VLA) models have recently been proposed as a pathway toward generalist robotic policies capable of interpreting natural language and visual inputs to generate manipulation actions. However, their effectiveness and efficiency on structured, long-horizon manipulation tasks rema…
Adapting quickly to dynamic, uncertain environments—often called “open worlds” —remains a major challenge in robotics. Traditional Task and Motion Planning (TAMP) approaches struggle to cope with unforeseen changes, are data-inefficient when adapting, and do not leverage world models during learning
Imitation learning enables intelligent systems to acquire complex behaviors with minimal supervision. However, existing methods often focus on short-horizon skills, require large datasets, and struggle to solve long-horizon tasks or generalize across task variations and distribution shifts. We propo…
Previous works have attempted to improve tracking efficiency through lightweight architecture design or knowledge distillation from teacher models to compact student trackers. However, these solutions often sacrifice accuracy for speed to a great extent, and also have the problems of complex trainin…
Time series forecasting is always important, with multivariate long-term series forecasting being its most challenging task. Here, the existing methods typically learn only in a single domain and focus on optimizing model structures, leading to incomplete information mining and imprecise predictions…
Natural language chain-of-thought (N-CoT) and Program chain-of-thought (P-CoT) have emerged as two primary paradigms for large language models (LLMs) to solve mathematical reasoning problems. Current research typically endeavors to achieve unidirectional enhancement: P-CoT enhanced N-CoT or N-CoT en
Vectorized high-definition (HD) map construction is an important and challenging task for autonomous driving. End-to-end models have been developed recently to enable online map construction. Existing works have difficulty on generating complex geometric shapes and lack comprehensive evaluation metr…
Recent advances in image deraining have focused on training powerful models on mixed multiple datasets comprising diverse rain types and backgrounds. However, this approach tends to overlook the inherent differences among rainy images, leading to suboptimal results. To overcome this limitation, we f…
Rain streaks will degrade the visibility of images. To tackle this problem, we propose a novel Adaptive Dilated Network (ADN) to remove rain streaks from a single image while using less parameters and running faster than previous methods. Specifically, an Adaptive Dilated Block (ADB) is constructed…
We describe a general-purpose end-to-end audio embeddings generator that can be easily adapted to various acoustic scene and event classification applications. In contrast to many other models for audio classification, this end-to-end embeddings generator does not require a separate feature extracti…
Recent attempts show that factorizing 3D convolutional filters into separate spatial and temporal components brings impressive improvement in action recognition. However, traditional temporal convolution operating along the temporal dimension will aggregate unrelated features, since the feature maps…