Leveraging large language models for automated chart summarization
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Access to graphical information on the internet remains a significant barrier for blind or visually impaired individuals, particularly when it comes to data visualizations like charts.This thesis explores how recent advancements in artificial intelligence can enhance accessibility through innovations in chart summarization, the process of automatically extracting information from a chart and compiling it into a textual summary intelligible to screen readers.The thesis focuses on two main areas of research.First, it applies the emerging architecture of large language model agents to the task of chart summarization, a novel application in this domain.This approach combines recent advancements in chart information extraction with the reasoning and planning capabilities of large language models.By leveraging natural language processing technologies, it reduces the need for curation of annotated datasets traditionally required to train vision AI models.An implementation of this agent-based approach is developed and evaluated, demonstrating its effectiveness in generating chart summaries.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it