docs: GLM-5.3-Flash attribution — wrong haraqat (10.05%), not convention (0.125pp) - #69
Merged
Merged
Conversation
…t convention Corrects the earlier driver claim. Per-position decomposition over the same 1,200 paragraphs, convention-normalized with U+0670 rules derived from the aligned positions (drop after ى, fatha elsewhere; controls r7/G-5.2 move <=0.009pp): GLM-5.3-Flash wrong-haraqat rate 10.05% vs GLM-5.2's 2.64% — which matches our r7 teacher's 2.62% to within 0.02pp — with missing 1.01%, extra 0.20%, and the entire dagger-alif convention effect at 0.125pp (310 marks, zero in GT). The regression is genuine mark errors at ~4x the predecessor rate; the Quranic marks explain the raw-mode skip flood, not the DER gap.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Corrects the driver claim in the GLM-5.3-Flash verdict and the results README. Data-derived attribution over the same 1,200 paragraphs: wrong-haraqat rate 10.05 percent vs GLM-5.2's 2.64 percent (itself matching our r7 teacher's 2.62 percent to within 0.02pp); missing 1.01, extra 0.20; the entire dagger-alif U+0670 convention effect is 0.125pp under rules derived from the aligned positions (drop after ى, fatha elsewhere) with r7/GLM-5.2 as <=0.009pp controls. The regression is genuine mark errors at ~4x the predecessor rate — the Quranic marks explain the raw-mode evaluator skips, not the DER gap.