REVIEW 4 cited by
Layout and Task Aware Instruction Prompt for Zero-shot Document Image Question Answering
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Layout-aware pre-trained models has achieved significant progress on document image question answering. They introduce extra learnable modules into existing language models to capture layout information within document images from text bounding box coordinates obtained by OCR tools. However, extra modules necessitate pre-training on extensive document images. This prevents these methods from directly utilizing off-the-shelf instruction-tuning language foundation models, which have recently shown promising potential in zero-shot learning. Instead, in this paper, we find that instruction-tuning language models like Claude and ChatGPT can understand layout by spaces and line breaks. Based on this observation, we propose the LAyout and Task aware Instruction Prompt (LATIN-Prompt), which consists of layout-aware document content and task-aware instruction. Specifically, the former uses appropriate spaces and line breaks to recover the layout information among text segments obtained by OCR tools, and the latter ensures that generated answers adhere to formatting requirements. Moreover, we propose the LAyout and Task aware Instruction Tuning (LATIN-Tuning) to improve the performance of small instruction-tuning models like Alpaca. Experimental results show that LATIN-Prompt enables zero-shot performance of Claude and ChatGPT to be comparable to the fine-tuning performance of SOTAs on document image question answering, and LATIN-Tuning enhances the zero-shot performance of Alpaca significantly. For example, LATIN-Prompt improves the performance of Claude and ChatGPT on DocVQA by 263% and 20% respectively. LATIN-Tuning improves the performance of Alpaca on DocVQA by 87.7%. Quantitative and qualitative analyses demonstrate the effectiveness of LATIN-Prompt and LATIN-Tuning. We provide the code in supplementary and will release it to facilitate future research.
Forward citations
Cited by 4 Pith papers
-
Doc-CoB: Enhancing Document Understanding with Visual Chain-of-Boxes Reasoning
A two-stage layout-focused visual reasoning method, Doc-CoB, improves document question answering by having the model select key layout boxes and then answer from those boxes.
-
DocVLM: Make Your VLM an Efficient Reader
DocVLM compresses OCR text and layout into 64 learned queries that are injected into frozen VLMs, boosting DocVQA accuracy at low image-token budgets across three model families.
-
InputSnatch: Stealing Input in LLM Services via Timing Side-Channel Attacks
A timing side-channel on shared LLM caches can partially reconstruct private user inputs in prompt-engineering and RAG services.
-
DLaVA: Document Language and Vision Assistant for Answer Localization with Enhanced Interpretability and Trustworthiness
By feeding a montage of text crops with numbered boxes to a vision-language model, DLaVA localizes answers in documents and reports stronger accuracy and localization than the baselines it compares against.
Discussion (0). Continue with ORCID to comment.