@@ -37,9 +39,8 @@ buvard is a **style linter**, not an AI detector. It finds statistically
...
@@ -37,9 +39,8 @@ buvard is a **style linter**, not an AI detector. It finds statistically
over-represented patterns; it cannot and does not claim to detect machine
over-represented patterns; it cannot and does not claim to detect machine
authorship. Its catalog is deliberately regex-shaped (high precision, low
authorship. Its catalog is deliberately regex-shaped (high precision, low
recall): règle de trois, synonyme acrobatique and sentence-length monotony
recall): règle de trois, synonyme acrobatique and sentence-length monotony
stay manual checks. docs/research.md holds the clustering study;
stay manual checks. See [Science and references](#science-and-references)
docs/research-litreview.md (in progress) holds the literature review and
for the evidence behind the catalog.
the falsifiable recommendations that will become catalog v3.
## Agents
## Agents
...
@@ -56,6 +57,34 @@ catalog as an agent skill: load it, run the linter, apply the findings by hand.
...
@@ -56,6 +57,34 @@ catalog as an agent skill: load it, run the linter, apply the findings by hand.
Hits are review candidates, never verdicts — the agent edits prose like an
Hits are review candidates, never verdicts — the agent edits prose like an
editor, not like a filter.
editor, not like a filter.
## Science and references
The catalog is data with provenance, and the barème is a ratified decision —
both are written down:
-**[docs/research.md](docs/research.md)** — the clustering study: five
anti-slop sources (fr1, fr2, étape, no-slop, ui-sk), what each contributes,
and how the shared patterns became the catalog.
-**[docs/research-litreview.md](docs/research-litreview.md)** — the
literature review on AI-text detection, and the **ten falsifiable
recommendations** that will become catalog v3. Read it before touching
the catalog.
Core citations, one line each — the lit review holds the full bibliography:
- Liang et al., 2023 — [GPT detectors are biased against non-native English writers](https://arxiv.org/abs/2304.02819): a 61.22% false-positive rate on TOEFL essays. The case for staying a linter.
-[MULTITuDE](https://arxiv.org/abs/2310.13606) — the large multilingual detection benchmark omits French; [INRIA's French ChatGPT-detection study](https://arxiv.org/abs/2306.05871) shows French is open territory.
-[GLTR](https://arxiv.org/abs/1906.04043), [DetectGPT](https://arxiv.org/abs/2301.11305), [Binoculars](https://arxiv.org/abs/2401.12070) — the detection toolbox; each degrades on short or paraphrased text.
- Sadasivan et al., 2024 — [paraphrase attacks collapse detection](https://arxiv.org/abs/2303.11156) on ~300-token passages. Why buvard never plays that game.
- Kobak et al., 2024 — [excess vocabulary](https://arxiv.org/abs/2403.07183) as a population-level AI marker. A cousin of our lexique.
- Sibling linters — [proselint](https://github.com/amperser/proselint) and [Vale](https://docs.vale.sh/) are the ancestors; [Wikipedia's "Signs of AI writing"](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing) is the shared folklore.
**Methodology, going forward (catalog v3):** every new rule must pass a
false-positive gate — under 2% per rule on a pre-2022 Gutenberg corpus,
human-written by construction. Model-generated text is used only to measure
sensitivity, never to admit a rule; a catalog tuned on one model's slop
learns that model's fingerprints.
## Skill
## Skill
`SKILL.md` ships the catalog as an agent skill: load it, run the linter,
`SKILL.md` ships the catalog as an agent skill: load it, run the linter,
<descid="d">One slop sentence dissected: buvard's catalog flags a binary contrast (−8), a servile promise (−10) and a second tiret cadratin beyond budget (−5); the score falls from 100 to 77.</desc>
<defs>
<style>
.ink { fill:#1f2a33; }
.rouge{ fill:#a63d2f; }
.muted{ fill:#7a6f5d; }
.serif{ font-family: Georgia, 'Times New Roman', serif; }
<textx="676"y="368"class="muted serif"font-size="13"font-style="italic">réécrire, puis re-passer</text>
</g>
<textx="440"y="458"class="sans muted"font-size="10"text-anchor="middle"letter-spacing="2">HITS ARE REVIEW CANDIDATES, NEVER VERDICTS — THE AUTHOR'S VOICE ALWAYS WINS</text>