Evaluating AI-Generated Emails: A Comparative Efficiency Analysis
Bibliographic record
Abstract
This study investigates the efficiency of large language models (LLMs) in producing routine, negative, and persuasive business emails for educational purposes within the context of Business Writing. Specifically, it compares the outputs generated by four widely-used LLMs (ChatGPT 3.5, Llama 2, Bing Chat, and Bard) when presented with identical email scenarios. These generated emails are evaluated using an elaborate rubric, allowing for a systematic assessment of LLMs' performance across three distinct email types. The results of the study show that the output with the same prompt varies greatly despite the rather formulaic nature of business emails. For instance, some LLMs struggle with following the requested structure and maintaining consistency in tone, while others have issues with unity and conciseness. The findings of this research hold implications for teaching business writing (rubrics, task instructions, in-class implementation), as well as for the integration of AI in professional communication at large.
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 | 0.049 | 0.411 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.008 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.001 |
| 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".