LLM Threat Query - Improving LLM KQL Generation; Bridging Actors And Accountability In AI Driven Misinformation Using Actor Network Theory; Earthquake Analysis And Prediction: Comparing ETAS And USGS Earthquake Data
Bibliographic record
Abstract
Technical Report Earthquakes pose a significant threat to human lives and infrastructure, providing a need for accurate earthquake forecasting models to help mitigate risk. My capstone research project presents a comprehensive analysis comparing synthetic earthquake simulations generated by the Epidemic-Type Aftershock Sequence (ETAS) model with historical earthquake records from the United States Geological Survey (USGS) spanning 1960 to 2023. I found discrepancies between modeled and observed data by analyzing patterns and energy related to significant earthquakes (magnitude ≥ 6.5). Notably, ETAS simulated data shows higher average energy levels before and after major earthquakes, while USGS data indicates a considerable energy spike just before the main event, which is not observed in ETAS data. STS Prospectus & Research Paper My STS research paper examines Air Canada’s 2024 chatbot failure through ANT and argues that the chatbot’s failure was a result of its actor-network rather than an isolated technical error. It examined several factors within Air Canada’s network that contributed to this failure including corporate strategy, industry trends, cheaper GPU access, and legal ambiguities and accountability. The technical project presented in the Prospectus develops a framework to improve the accuracy of Large Language Model (LLM) queries in generating Kusto Query Language (KQL) queries. Working on both projects simultaneously enhanced my understanding of AI products and failures in their implementation. The technical project allowed me to focus on combating AI hallucinations from a technical perspective, enriched by insights from STS research, which shows how socio-technical dynamics shape technical design. Note: technical project referenced is LLM Threat Query that I have been working on during fall 2024 - spr 2025 which is unrelated to my capstone research - Earthquake Analysis And Prediction, that I completed during spr 2024.
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 | 0.010 | 0.045 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.007 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".