Leo Leppänen
@ljleppan.bsky.social
Postdoctoral researcher in computer science at University of Helsinki MOOC Center. Interested in automated journalism, natural language generation, learning analytics, computer science education.
"Foster" was another word that increased 10-fold (10.84x, to be exact) in prevalence when comparing pre- and post-ChatGPT data.
We also noted massive jumps in mean answer length (both tokens and sentences), Flesch Reading Ease and the mean Type-Token Ratio starting 3mo post-ChatGPT.
We also noted massive jumps in mean answer length (both tokens and sentences), Flesch Reading Ease and the mean Type-Token Ratio starting 3mo post-ChatGPT.
June 14, 2025 at 7:12 AM
"Foster" was another word that increased 10-fold (10.84x, to be exact) in prevalence when comparing pre- and post-ChatGPT data.
We also noted massive jumps in mean answer length (both tokens and sentences), Flesch Reading Ease and the mean Type-Token Ratio starting 3mo post-ChatGPT.
We also noted massive jumps in mean answer length (both tokens and sentences), Flesch Reading Ease and the mean Type-Token Ratio starting 3mo post-ChatGPT.
Kobak et al. (2025, arxiv.org/pdf/2406.07016) analyzed pubmed abstract from 2010-2024, identifying a bunch of words that are at least highly suspicious.
In upcoming work for AIED'25, we looked at student MOOC essays from 2020-2024 and identified very similar trends, e.g. "delve" increasing 10-fold.
In upcoming work for AIED'25, we looked at student MOOC essays from 2020-2024 and identified very similar trends, e.g. "delve" increasing 10-fold.
arxiv.org
June 14, 2025 at 6:58 AM
Kobak et al. (2025, arxiv.org/pdf/2406.07016) analyzed pubmed abstract from 2010-2024, identifying a bunch of words that are at least highly suspicious.
In upcoming work for AIED'25, we looked at student MOOC essays from 2020-2024 and identified very similar trends, e.g. "delve" increasing 10-fold.
In upcoming work for AIED'25, we looked at student MOOC essays from 2020-2024 and identified very similar trends, e.g. "delve" increasing 10-fold.
Reposted by Leo Leppänen
/3 But here’s the great thing, the vendor explained: you can tell the AI not to include hallucinations in its work product.
I beg your pardon, I said to the paralegal relating this to me.
You can set hallucinations to “no,” he explained.
I beg your pardon, I said to the paralegal relating this to me.
You can set hallucinations to “no,” he explained.
November 29, 2023 at 8:24 PM
/3 But here’s the great thing, the vendor explained: you can tell the AI not to include hallucinations in its work product.
I beg your pardon, I said to the paralegal relating this to me.
You can set hallucinations to “no,” he explained.
I beg your pardon, I said to the paralegal relating this to me.
You can set hallucinations to “no,” he explained.