From 7923eda4abea4fcaea82a646354b23f1a6ce602d Mon Sep 17 00:00:00 2001 From: Ronald Tse Date: Mon, 7 Sep 2026 08:47:06 +0200 Subject: [PATCH] docs: README and paper-a point at index-v5 The README's artifact table still said index-v4 (one release behind after the fp16 entry cut v5); paper-a's artifact paragraph said the same. Both now say v5 - the index that resolves the full 23-model catalog. --- README.md | 2 +- docs/paper-a.adoc | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 398191f..f54f5a6 100644 --- a/README.md +++ b/README.md @@ -25,7 +25,7 @@ This repo owns three things and nothing else: | Protocol CLI | `pip install interscript-ml-tools[sadeed]` ([PyPI](https://pypi.org/project/interscript-ml-tools/)) | | Raw predictions, every frontier run | Release [`frontier-predictions-v1`](https://github.com/interscript/interscript-ml/releases/tag/frontier-predictions-v1) | | Teachers + provenance labels | Release [`teachers-arabic-v1`](https://github.com/interscript/interscript-ml/releases/tag/teachers-arabic-v1) | -| Student artifacts | `models.yaml` (index-v4) — sha256-verified in every runtime | +| Student artifacts | `models.yaml` (index-v5) — sha256-verified in every runtime | interscript-sadeed-eval score --preds --data Misraj/SadeedDiac-25 diff --git a/docs/paper-a.adoc b/docs/paper-a.adoc index db7f2a0..c9d3363 100644 --- a/docs/paper-a.adoc +++ b/docs/paper-a.adoc @@ -263,7 +263,7 @@ DER-CE scoring protocol as the `interscript-sadeed-eval` CLI (PyPI: raw predictions and the public benchmark; the full measurement ledger (RESULTS.md) with every verdict, interval, and provenance hash. Paper B carries the client-tier artifacts (distilled IMF v1 zips resolvable -from index-v4, parity and margin reports, cross-runtime byte parity). +from index-v5, parity and margin reports, cross-runtime byte parity). == References