Skip to content
C.W.K.
Stream
Lesson 04 of 05 · published

Fix the Lens, Not the Vault

~12 min · false-positive, detector, regression, measurement

Level 0Trace
0 XP0/36 lessons0/12 achievements
0/100 XP to next level100 XP to go0% complete

The temptation to satisfy the proxy

When a detector flags valid prose, the fastest green is a small rewrite. The check passes and the dashboard quiets, but source has changed to match an instrument bug rather than meaning.

This is especially dangerous for memory. Removing placeholders, names, or intentional repetition that the lens cannot model deprives the next soul of information. The test passes while the product purpose fails.

Verify the premise first

Read the candidate and decide whether the claimed class is actually present. On a false positive, restore content and narrow the predicate. Prefer an exact description of the legitimate shape over a broad ignore.

If an English X triggers a Korean shorthand detector, allowing every X hides real defects. Classify placeholder position, code block, or mathematical symbol and exempt only those contexts.

Keep the incident shape as a fixture

After repair, preserve a reduced version of the real false-positive fragment. The valid sample must pass while a neighboring invalid sample still fails. With only one side, disabling the detector also appears green.

Place samples on both sides of numeric thresholds and retain raw syntax for parser failures. Generalize the lesson without erasing the unusual shape that exposed it.

State instrument limits in the UI

No detector is perfect. Candidate cards should show why they were flagged and known limitations so users read the result as evidence rather than an oracle.

When reporting false-positive rate, attach period and sample. A clean run does not prove perfect precision; show how many reviewed candidates were confirmed.

Source does not exist to please a detector; the detector exists to read source better. A false positive is evidence to repair the instrument, not a reason to alter memory.

Code

Fix valid and invalid cases together·python
def suspicious_x(text, in_code=False):
    if in_code:
        return False
    return " X " in f" {text} " and "placeholder X" not in text

assert not suspicious_x("replace placeholder X with a value")
assert not suspicious_x("x = 3", in_code=True)
assert suspicious_x("이 단계는 X 하지 마")

External links

Exercise

Directly classify ten recent detector hits, calculate precision, and turn one false positive into a valid/invalid regression pair.
Hint
Keep the minimum context that explains why the false positive is valid.

Progress

Progress is local-only — sign in to sync across devices.
Spotted a bug or have feedback on this page?Report an Issue

Comments 0

🔔 Reply notifications (sign in)
Sign inPlease sign in to comment.

No comments yet — be the first.