18 accepted papers
Reinforcement learning (RL) has advanced LLM agents on verifiable tasks but remains challenging for open-ended tasks with vast solution spaces (e.g., complex travel planning). Lacking objective ground truth, current RL algorithms rely on reward models assigning scalar scores to individual responses.…
3D multi-object tracking is a critical and challenging task in the field of autonomous driving. A common paradigm relies on modeling individual object motion, e.g., Kalman filters, to predict trajectories. While effective in simple scenarios, this approach often struggles in crowded environments or
Implicit Neural Representation (INR) has emerged as a powerful paradigm for continuous MRI reconstruction. However, standard unsupervised INR requires time-consuming optimization from scratch for each scan, hindering clinical deployment. This work presents IPOD, a Reference-Free Meta-Learning framew…
Video Large Language Models (VLLMs) demonstrate strong video understanding but suffer from inefficiency due to redundant visual tokens. Existing pruning primary targets intra-frame spatial redundancy or prunes inside the LLM with shallow-layer overhead, yielding suboptimal spatiotemporal reduction a
Harnessing the event-driven characteristic, Spiking Neural Networks (SNNs) present a promising avenue toward energy-efficient Transformer architectures. However, existing Spiking Transformers still suffer significant performance gaps compared to their Artificial Neural Network counterparts. Through…
LiDAR-based 3D object detection is widely used in high-level autonomous driving schemes. However, the cumbersome modules in most 3D detectors lead to substantial computational overhead. Despite knowledge distillation (KD) is an effective approach for compressing models, previous methods cannot be ex…
Low probability of intercept (LPI) radar signals are widely used in modern electromagnetic warfare due to their exceptional anti-interception capabilities. A defining characteristic of LPI radar signals is their low peak power, which makes them highly susceptible to being masked by additive white Ga…
The development of Multimodal Large Language Models (MLLMs) has seen significant progress, driven by increasing demands across various fields (e.g., multimodal agents, embodied intelligence). While model-driven approaches aim to enhance MLLM capabilities through diverse architectures, their performa…
Role-Playing Agents (RPAs), benefiting from large language models, is an emerging interactive AI system that simulates roles or characters with diverse personalities. However, existing methods primarily focus on mimicking dialogues among roles in textual form, neglecting the role’s voice traits (e.g…
Recent advancements in omnimodal learning have significantly improved understanding and generation across images, text, and speech, yet these developments remain predominantly confined to proprietary models. The lack of high-quality omnimodal datasets and the challenges of real-time emotional speech…
Despite the widespread use of Transformer-based text embedding models in NLP tasks, surprising “sticky tokens” can undermine the reliability of embeddings. These tokens, when repeatedly inserted into sentences, pull sentence similarity toward a certain value, disrupting the normal distribution of em…
Knowledge distillation (KD) possesses immense potential to accelerate the deep neural networks (DNNs) for LiDAR-based 3D detection. However in most of prevailing approaches the suboptimal teacher models and insufficient student architecture investigations limit the performance gains. To address thes…
Among existing Neural Architecture Search methods, DARTS is known for its efficiency and simplicity. This approach applies continuous relaxation of network representation to construct a weight-sharing supernet and enables the identification of excellent subnets in just a few GPU days. However, perfo…
User review data is considered as auxiliary information to alleviate the data sparsity problem and improve the quality of learned user/item or interaction representations in review-based recommender systems. However, existing methods usually model user-item interactions in a holistic manner and negl…
Cross-modal retrieval aims to build correspondence between multiple modalities by learning a common representation space. Typically, an image can match multiple texts semantically and vice versa, which significantly increases the difficulty of this task. To address this problem, probabilistic embedd…
Transformer-based architectures with grid features represent the state-of-the-art in visual and language reasoning tasks, such as visual question answering and image-text matching. However, directly applying them to image captioning may result in spatial and fine-grained semantic information loss. T…
In this paper, we aim to recognize materials with combined use of auditory and visual perception. To this end, we construct a new dataset named GLAudio that consists of both the geometry of the object being struck and the sound captured from either modal sound synthesis (for virtual objects) or real…