A5063725719- Memory and Neural Mechanisms
- Neural dynamics and brain function
- Natural Language Processing Techniques
- Neurobiology and Insect Physiology Research
- Advanced Graph Neural Networks
- Human Mobility and Location-Based Analysis
- Neural and Behavioral Psychology Studies
- Topic Modeling
- Consumer Perception and Purchasing Behavior
- Complex Network Analysis Techniques
- Speech Recognition and Synthesis
The University of Tokyo
2024
Multimodal large language models (MLLMs) have garnered widespread attention due to their ability understand multimodal input. However, parameter sizes and substantial computational demands severely hinder practical deployment application.While quantization is an effective way reduce model size inference latency, its application MLLMs remains underexplored. In this paper, we propose MQuant, a post-training (PTQ) framework designed tackle the unique challenges of (MLLMs). Conventional often...
Abstract When a simple model-free strategy does not provide sufficient outcomes, an inference-based estimating hidden task structure becomes essential for optimizing choices. However, the neural circuitry involved in strategies is still unclear. We developed tone frequency discrimination head-fixed mice which category of current trial depended on previous trial. was repeated every trial, continued to use default strategy, as well when randomly presented, bias In contrast, gradually shifted...
Community detection is the task of discovering groups nodes sharing similar patterns within a network. With recent advancements in deep learning, methods utilizing graph representation learning and clustering have shown great results community detection. However, these often rely on topology networks (i) ignoring important features such as network heterogeneity, temporality, multimodality, other possibly relevant features. Besides, (ii) number communities not known priori left to model...