Deep Learning
LIVE: Learnable In-Context Vector for Visual Question Answering
As language models continue to scale, Large Language Models (LLMs) have exhibited emerging capabilities in In-Context Learning (ICL), enabling them to solve language tasks by prefixing a few in-context demonstrations (ICDs) as context. Inspired by these advancements, researchers have extended these techniques to develop Large Multimodal Models (LMMs) with ICL capabilities.
HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning
B matrices to avoid training interference.
PrivAuditor: Benchmarking Privacy Vulnerabilities in LLM Adaptation Techniques
Despite these achievements, LLMs often underperform in domain-specific tasks without training on relevant domain data.
UKnow: A Unified Knowledge Protocol with Multimodal Knowledge Graph Datasets for Reasoning and Vision-Language Pre-Training
See Appendix A to download the dataset.
Meta 3D AssetGen: T ext-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials
Compared to works that bake shading in the 3D object's appearance,
Unlocking Tokens as Data Points for Generalization Bounds on Larger Language Models
With Monarch matrices, Kronecker factorizations, and post-training quantization, we achieve non-vacuous generalization bounds for LLMs as large as LLaMA2-70B.