bavitutor 0.2.0 → 0.3.1

package/README.md

@@ -135,22 +135,32 @@ The three `BAVI_MODEL_*` variables are only starting points: **Settings → AI m
    Reviews interleave subjects, pick the question level from memory strength
    (one lower after a lapse, discrimination items against confusable
    concepts), compress overlapping concepts into fewer items, and give
-   implicit credit to prerequisites. A hint ladder caps how much a helped
-   answer counts. Concepts that keep slipping get flagged with remediation;
-   a backlog switches to triaged catch-up spread over a week; after a long
-   absence a 5-minute check-in replaces the pile. "Just 5" keeps it short.
+   implicit credit to prerequisites. You never rate yourself: "I don't know"
+   is the honest no, and each rating comes from whether you were right, how
+   long you took against your usual time, and any hints. A hint ladder caps
+   how much a helped answer counts. Anything you miss or struggle with comes
+   back later in the session, as a new question, until you get it right
+   without a struggle. A session ends by naming the concepts that were
+   strong and the ones that need practice. Concepts that keep slipping get
+   flagged with remediation; a backlog switches to triaged catch-up spread
+   over a week; after a long absence a 5-minute check-in replaces the pile.
+   "Just 5" keeps it short.
    Every concept is also split into a few **ideas**, each with its own FSRS
    card: a flashcard round closes each lesson, and due ideas come back as
    flashcards written fresh each time from how they were asked before. You
    recall the answer first, tap "Show choices", then pick the right statement
    out of four near-identical ones; recall time and the pick set the rating.
+   A concept only counts as Mastered once most of its ideas are held, ideas
+   that keep slipping get a new lesson aimed at them, a missed card can be
+   explained again from another angle, lessons you skip still get their
+   ideas drilled, and Progress shows how many ideas you can recall.
 6. **Your day.** Home plans reviews first, then as many new lessons as fit
    your daily time and a two-week review-load forecast. Settings cover daily
    time, retention, new-concept cap, visuals, a daily reminder, and the
    Claude model, reasoning level and fast mode for each kind of work. A goal
    can have an exam date that tightens scheduling. Progress shows durable
-   knowledge, predicted and measured retention, calibration, activity and a
-   review forecast. Lessons resume after a reload, and a break is suggested
+   knowledge, predicted and measured retention, activity and a review
+   forecast. Lessons resume after a reload, and a break is suggested
    after about 25 minutes.
 
 ### Learning from your own sources