explore: more variance in the related rail (stronger MMR diversification) #176

2026-07-01T00:50:54-04:00

bvandeusen commented

2026-07-01 00:50:54 -04:00

The Explore "related" rail (gallery_service.similar → _diversify_similar, from #1188) was tuned conservatively. Push all three diversity knobs so it spans further across clusters instead of clumping near the anchor:

MMR lam 0.55 → 0.40 — weight the diversity penalty harder (the main dial).
candidate pool min(200, max(limit*5, 60)) → min(400, max(limit*8, 100)) — a wider nearest-cosine pool so MMR has genuinely distinct neighbourhoods to reach into.
pHash dup_threshold 6 → 8 — collapse more near-duplicate reposts/clones, freeing rail slots for distinct picks.

Still deterministic (same set per image, just more spread) and relevance-anchored via the lam*sim-to-anchor term. Backend-only, no migration. CI green on dev (run 1818, incl. integration).

🤖 Generated with Claude Code

The Explore "related" rail (`gallery_service.similar` → `_diversify_similar`, from #1188) was tuned conservatively. Push all three diversity knobs so it spans further across clusters instead of clumping near the anchor: - **MMR `lam` 0.55 → 0.40** — weight the diversity penalty harder (the main dial). - **candidate pool `min(200, max(limit*5, 60))` → `min(400, max(limit*8, 100))`** — a wider nearest-cosine pool so MMR has genuinely distinct neighbourhoods to reach into. - **pHash `dup_threshold` 6 → 8** — collapse more near-duplicate reposts/clones, freeing rail slots for distinct picks. Still deterministic (same set per image, just more spread) and relevance-anchored via the `lam*sim-to-anchor` term. Backend-only, no migration. CI green on dev (run 1818, incl. integration). 🤖 Generated with [Claude Code](https://claude.com/claude-code)

bvandeusen added 1 commit 2026-07-01 00:50:54 -04:00

feat(explore): more variance in the related rail (stronger MMR diversification)

CI / lint (push) Successful in 3s

Details

CI / frontend-build (push) Successful in 21s

Details

CI / backend-lint-and-test (push) Successful in 26s

Details

CI / integration (push) Successful in 3m24s

Details

ef3318aac1

Operator wants the Explore "related" rail to span more — the #1188 diversifier
was tuned conservatively. Push all three knobs so it reaches further across
clusters instead of clumping near the anchor:

- MMR lam 0.55 → 0.40 — weight the diversity penalty harder (the main dial).
- candidate pool min(200, max(limit*5, 60)) → min(400, max(limit*8, 100)) — a
  wider nearest-cosine pool so MMR has genuinely distinct neighbourhoods to pick
  from, not just the near-dupes.
- pHash dup_threshold 6 → 8 — collapse more near-duplicate reposts/clones,
  freeing rail slots for distinct picks.

Still deterministic (same set per image, just more spread) and relevance-anchored
via the lam*sim-to-anchor term. Backend-only; no migration.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa

bvandeusen merged commit a78f7eaace into main

2026-07-01 00:50:59 -04:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: bvandeusen/FabledCurator#176