RAG-Enhanced Safety Information Retrieval for Construction: Integration of Large Language Models with Domain-Specific Information
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
In the construction industry, critical safety information is often scattered across numerous documents, standards, and regulations, making it challenging for practitioners to access and comprehend safety knowledge in their daily operations efficiently.To address this challenge, we propose an intelligent and reliable questionanswering system for information retrieval and response generation on the construction health, safety, and environment documents via retrieval-augmented generation.Specifically, our system combines a finetuned LLaMA-3-8B base model with a vector database constructed using embedding models, enabling accurate information retrieval and enhancing the generated responses' reliability.Initial validation using cosine similarity analysis demonstrates promising results, with our system achieving a cosine similarity score of 0.936, outperforming the LLAMA3-8B base model's score of 0.884 in processing construction safety documentation.The preliminary findings show that: 1) our RAG-enhanced system provides safety information access, and 2) our specialized preprocessing techniques effectively synthesize and retrieve safety information, reducing fragmentation and access time.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.002 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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