Records found: 3
'Baidu launches ERNIE-4.5-VL-28B-A3B-Thinking, a compact open-source multimodal model that activates 3B parameters per token while offering strong document, chart and video reasoning capabilities.'
TableRAG is a novel hybrid AI framework that integrates SQL and text retrieval to improve multi-hop question answering over documents containing both tables and text, demonstrating superior performance on complex datasets.
NVIDIA releases Llama Nemotron Nano VL, a compact vision-language model that excels in complex document understanding with efficient multimodal processing and state-of-the-art accuracy.