here<\/a>. It presents a fully reproducible, embedding-based analysis of the PhD theses by Arday (2015) and Zwozdiak-Myers (2009).<\/p>\n\n\n\nWonderfully, the complete pipeline is included end-to-end: extraction, segmentation, embedding, control construction, Poisson testing, provenance classification, and extent estimation. The code automatically retrieves all source documents from their respective repositories. Equally remarkably, the entire process runs on a standard CPU in approximately twenty minutes.<\/p>\n\n\n\n
The two PhD theses have considerable overlap.<\/strong><\/p>\n\n\n\n\n“The signal is in the tail, and it survives controls. Against two independent education doctorates matched on discipline and partially on topic, the focal pair shows:<\/p>\n\n\n\n
\n- 9 near-identical sentence pairs (cosine \u2265 0.999) in 11.09M comparisons, against 1<\/strong> in 34.29M control comparisons – and that single control match is the standard thesis declaration (\u201cThis work has not previously been presented for an award at this, or any other, university\u201d).<\/li>\n\n\n\n
- 179 sentence pairs sharing 12 or more consecutive identical words (107 spatially independent blocks), against an expected 0.32 under the null – an exact count over all 11.09M pairs, no embedding model involved. Longest shared run: 63 words.<\/li>\n\n\n\n
- Conditional on high semantic similarity, cross-document pairs share a median run of 11.5 identical consecutive words, versus 4.0 and 7.5 for sentences within<\/em> each thesis – that is, among semantically matched sentences, the two documents phrase shared ideas alike more often than either thesis does when restating itself.<\/li>\n<\/ul>\n\n\n\n
Under the most conservative assumptions available, the probability of observing this pattern under independent composition is 5.7 \u00d7 10\u207b\u2074; on the central estimate of the null rate, 8 \u00d7 10\u207b\u00b9\u00b9. “<\/p>\n\n\n\n
[From Titorekno’s GitHub page]<\/em><\/p>\n<\/blockquote>\n\n\n\nIt is very unlikely this happened independently (specifically one in a one-in-100-billion chance of happening).<\/p>\n\n\n\n
The origin of the extreme textual overlap perhaps remains unclear — but both Liverpool John Moores University and Cambridge University have got to come up with better explanations than they have done so far.<\/p>\n\n\n\n
And for Cambridge University’s senior management, there are serious questions as to why this is playing out so damagingly in public, since they have known about it for a number of years.<\/p>\n","protected":false},"excerpt":{"rendered":"
There is now a lot of press coverage of in The Times, the Telegraph and the Mail. From a mathematician’s or programmer’s perspective, we highlight an outstanding piece of work by Alexey Titorenko, available to download here. It presents a fully reproducible,\u2026<\/p>\n","protected":false},"author":1,"featured_media":4037,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-4036","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-blog"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t