Computer Science & Information Technology (CS & IT)
Bibliographic record
Abstract
CAPTCHA is almost a standard security technology, and has found widespread application in commercial websites.There are two types: labeling and image based CAPTCHAs.To date, almost all CAPTCHA designs are labeling based.Labeling based CAPTCHAs refer to those that make judgment based on whether the question "what is it?"has been correctly answered.Essentially in Artificial Intelligence (AI), this means judgment depends on whether the new label provided by the user side matches the label already known to the server.Labeling based CAPTCHA designs have some common weaknesses that can be taken advantage of attackers.First, the label set, i.e., the number of classes, is small and fixed.Due to deformation and noise in CAPTCHAs, the classes have to be further reduced to avoid confusion.Second, clean segmentation in current design, in particular character labeling based CAPTCHAs, is feasible.The state of the art of CAPTCHA design suggests that the robustness of character labeling schemes should rely on the difficulty of finding where the character is (segmentation), rather than which character it is (recognition).However, the shapes of alphabet letters and numbers have very limited geometry characteristics that can be used by humans to tell them yet are also easy to be indistinct.Image recognition CAPTCHAs faces many potential problems which have not been fully studied.It is difficult for a small site to acquire a large dictionary of images which an attacker does not have access to and without a means of automatically acquiring new labeled images, an image based challenge does not usually meet the definition of a CAPTCHA.They are either unusable or prone to attacks.In this paper, we present the different types of CAPTCHAs trying to defeat advanced computer programs or bots, discussing the limitations and drawbacks of each.
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 | 0.001 | 0.007 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.267 | 0.261 |
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".