With the increasing adoption of Large Language Models (LLMs) and Vision-Language Models (VLMs), rich document analysis technologies for applications like Retrieval-Augmented Generation (RAG) and visual RAG are gaining significant attention. Recent research indicates that using VLMs yields better RAG performance, but processing rich documents remains a challenge since a single page contains large amounts of information. In this paper, we present SCAN (SemantiC Document Layout ANalysis), a novel approach that enhances both textual and visual Retrieval-Augmented Generation (RAG) systems that work with visually rich documents. It is a VLM-friendly approach that identifies document components with appropriate semantic granularity, balancing context preservation with processing efficiency. SCAN uses a coarse-grained semantic approach that divides documents into coherent regions covering contiguous components. We trained the SCAN model by fine-tuning object detection models on an annotated dataset. Our experimental results across English and Japanese datasets demonstrate that applying SCAN improves end-to-end textual RAG performance by up to 9.4 points and visual RAG performance by up to 10.4 points, outperforming conventional approaches and even commercial document processing solutions.

In this paper, we introduce a novel method with graph-to-segment mapping for question answering over knowledge graphs, which helps understanding question utterances. This method centers on semantic parsing, a key approach for interpreting these utterances. The challenges lie in comprehending implicit entities, relationships, and complex constraints like time, ordinality, and aggregation within questions, contextualized by the knowledge graph. Our framework employs a combination of rule-based and neural-based techniques to parse and construct highly accurate and comprehensive semantic segment sequences. These sequences form semantic query graphs, effectively representing question utterances. We approach question semantic parsing as a sequence generation task, utilizing an encoder-decoder neural network to transform natural language questions into semantic segments. Moreover, to enhance the parsing of implicit entities and relations, we incorporate a graph neural network that leverages the context of the knowledge graph to better understand question representations. Our experimental evaluations on two datasets demonstrate the effectiveness and superior performance of our model in semantic parsing for question answering.

This paper proposes a method of abstractive summarization designed to scale to document collections instead of individual documents. Our approach applies a combination of semantic clustering, document size reduction within topic clusters, semantic chunking of a cluster's documents, GPT-based summarization and concatenation, and a combined sentiment and text visualization of each topic to support exploratory data analysis. Statistical comparison of our results to existing state-of-the-art systems BART, BRIO, PEGASUS, and MoCa using ROGUE summary scores showed statistically equivalent performance with BART and PEGASUS on the CNN/Daily Mail test dataset, and with BART on the Gigaword test dataset. This finding is promising since we view document collection summarization as more challenging than individual document summarization. We conclude with a discussion of how issues of scale are