Work-in-Progress: Fine-Tuning Large Language Models for Automated Feedback in Complex Engineering Problem-Solving
Bibliographic record
Abstract
This paper presents work in progress (WIP) toward using artificial intelligence (AI), specifically through Large Language Models (LLM), to support rapid quality feedback mechanisms within engineering educational settings.It describes applying to LLMs to improve the feedback processes by providing information directly to students, graders, or course instructors teaching courses focused on complex engineering problem-solving.We detail how fine-tuning an LLM with a small dataset from diverse problem scenarios achieves classification accuracies close to approximately 80%, even in new problems not included in the fine-tuning process.Traditionally, open-source LLMs, like BERT, have been fine-tuned in large datasets for specific domain tasks.Our results suggest this may not be as critical in achieving good performances as previously thought.Our findings demonstrated the potential for applying AI-supported personalized feedback through high-level prompts incentivizing students to critically self-assess their problem-solving process and communication.However, this study also highlights the need for further research into how semantic diversity and synthetic data augmentation can optimize training datasets and impact model performance.
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 | 0.007 | 0.030 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".