Recent advances in multimodal retrieval-augmented generation (MM-RAG) have shifted toward minimal parsing, relying on page-level images for producing retriever embeddings and for answer generation. While efficient, this trend often neglects explicit handling of the rich, structured information in complex enterprise documents, instead depending on pre-trained embeddings or vision-language models to implicitly capture such structure. In this work, we take a more direct approach: MM-BizRAG proactively extracts and represents document structure via a document structure-aware split that dynamically routes documents through orientation-specific ingestion pipelines, applying explicit layout-aware parsing for vertically structured documents (e.g., reports) and holistic page-level representations for horizontally structured documents (e.g., slide decks). A unified LLM-driven artifact transformation pipeline with placeholder-based positional alignment preserves natural reading order, while inference-time multimodal assembly decouples retrieval representations from generation context, enabling richer, more grounded answers without any finetuning requirement. Through experiments on a large, heterogeneous enterprise dataset and two public benchmarks (SlideVQA and FinRAGBench-V), MM-BizRAG consistently outperforms state-of-the-art vision-centric baselines by up to 32% points, with especially strong gains on report-style layouts. Furthermore, we introduce FastRAGEval, a single-call LLM Judge metric for fine-grained generative recall that halves RAGChecker’s cost while achieving stronger human alignment.

We present a large language model (LLM) based approach for comparing legal contracts with their corresponding template documents. Legal professionals use commonly observed deviations between templates and contracts to help with contract negotiations, and also to refine the template documents. Our comparison approach, based on the well-studied natural language inference (NLI) task, first splits a template into key concepts and then uses LLMs to decide if the concepts are entailed by the contract document. We also repeat this procedure in the opposite direction - contract clauses are tested for entailment against the template clause to see if they contain additional information. The non-entailed concepts are labelled, organized and filtered by frequency, and placed into a clause library, which is used to suggest changes to the template documents. We first show that our LLM-based approach outperforms all previous work on a publicly available dataset designed for NLI in the legal domain. We then apply it to a private real-world legal dataset, achieve an accuracy of 96.46%. Our approach is the first in the literature to produce a natural language comparison between legal contracts and their template documents.

This paper applies contextualized word embedding models to a long-standing problem in the natural language parsing community, namely prepositional phrase attachment. Following past formulations of this problem, we use data sets in which the attachment decision is both a binary-valued choice as well as a multi-valued choice. We present a deep learning architecture that fine-tunes the output of a contextualized word embedding model for the purpose of predicting attachment decisions. We present experiments on two commonly used datasets that outperform the previous best results, using only the original training data and the unannotated full sentence context.

Athena 2.0 is an Alexa Prize SocialBot that has been a finalist in the last two Alexa Prize Grand Challenges. One reason for Athena’s success is its novel dialogue management strategy, which allows it to dynamically construct dialogues and responses from component modules, leading to novel conversations with every interaction. Here we describe Athena’s system design and performance in the Alexa Prize during the 20/21 competition. A live demo of Athena as well as video recordings will provoke discussion on the state of the art in conversational AI.