Large language models for closed-library multi-document query, test generation, and evaluation
Bibliographic record
Abstract
Introduction: Learning complex, detailed, and evolving knowledge is a challenge in multiple technical professions. Relevant source knowledge is contained within many large documents and information sources with frequent updates to these documents. Knowledge tests need to be generated on new material and existing tests revised, tracking knowledge base updates. Large Language Models (LLMs) provide a framework for artificial intelligence-assisted knowledge acquisition and continued learning. Retrieval-Augmented Generation (RAG) provides a framework to leverage available, trained LLMs combined with technical area-specific knowledge bases. Methods: Herein, two methods are introduced (DaaDy: document as a dictionary and SQAD: structured question answer dictionary), which together enable effective implementation of LLM-RAG question-answering on large documents. Additionally, the AI for knowledge intensive tasks (AIKIT) solution is presented for working with numerous documents for training and continuing education. AIKIT is provided as a containerized open source solution that deploys on standalone, high performance, and cloud systems. AIKIT includes LLM, RAG, vector stores, relational database, and a Ruby on Rails web interface. Results: Coverage of source documents by LLM-RAG generated questions decreases as the length of documents increase. Segmenting source documents improve coverage of generated questions. The AIKIT solution enabled easy use of multiple LLM models with multimodal RAG source documents; AIKIT retains LLM-RAG responses for queries against one or multiple LLM models. Discussion: AIKIT provides an easy-to-use set of tools to enable users to work with complex information using LLM-RAG capabilities. AIKIT enables easy use of multiple LLM models with retention of LLM-RAG responses.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.015 | 0.050 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.013 | 0.006 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".