Search Results for author:
Found 11 papers, 3 papers with code
VDocRAG: Retrieval-Augmented Generation over Visually-Rich Documents
We aim to develop a retrieval-augmented generation (RAG) framework that answers questions over a corpus of visually-rich documents presented in mixed modalities (e. g., charts, tables) and diverse formats (e. g., PDF, PPTX).
Wavelet-based Positional Representation for Long Context
From these insights, we propose a new position representation method that captures multiple scales (i. e., window sizes) by leveraging wavelet transforms without limiting the model's attention field.
ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind
These verbalized thoughts serve as answers to questions designed to assess the mental states of characters within conversations.
Initialization of Large Language Models via Reparameterization to Mitigate Loss Spikes
WeSaR introduces a gate parameter per parameter matrix and adjusts it to the value satisfying the requirements.
InstructDoc: A Dataset for Zero-Shot Generalization of Visual Document Understanding with Instructions
We study the problem of completing various visual document understanding (VDU) tasks, e. g., question answering and information extraction, on real-world documents through human-written instructions.
SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images
Visual question answering on document images that contain textual, visual, and layout information, called document VQA, has received much attention recently.
Automatically Extracting Variant-Normalization Pairs for Japanese Text Normalization
Social media texts, such as tweets from Twitter, contain many types of non-standard tokens, and the number of normalization approaches for handling such noisy text has been increasing.
Constructing a Class-Based Lexical Dictionary using Interactive Topic Models
This paper proposes a new method of constructing arbitrary class-based related word dictionaries on interactive topic models; we assume that each class is described by a topic.
