Restore Gemini 3.1 Pro data in Paper 1#4
Open
byapparov wants to merge 4 commits into
Open
Conversation
Revert main.tex to pre-removal version (1e1f15b) with all 5-model/737-run/3-family numbers. Update compute_stats.py and generate_figures.py to load Gemini data from papers/3-kpi-targets/gemini_scores.csv. Regenerate all figures and PDF. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Soften 'identical semantic content' claim to acknowledge pseudocode's structural priors (typed fields, enums, validation functions). Add mechanistic hypothesis for Gemini SQL anomaly (over-literal PerFileRules interpretation). Expand future work with ablation studies and hybrid format suggestions. Update threats to validity accordingly. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add 4 new verified citations: Willard & Louf (constrained decoding), Liu et al. (Text-to-SQL survey), Lyu et al. (Dockerfile generation, ICSE 2026), Madaan et al. (Self-Refine, NeurIPS 2023) - Integrate citations into Related Work (Code as Prompt, new Constrained Decoding paragraph) and Discussion (Relation to constrained decoding) - Fix Table 2 column overflow by shortening task names and header - Add metadata reminder comment for ACM rights management - Add QA.md documenting run independence analysis for reviewer questions Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Use relative paths with a common prefix note instead of full
repository paths to keep \texttt{} blocks within column width.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
main.texto pre-removal version (commit1e1f15b) restoring all 5-model / 737-run / 3-family numbers, Gemini table rows, and discussion textcompute_stats.pyandgenerate_figures.pyto load Gemini data frompapers/3-kpi-targets/gemini_scores.csv(withsql-query→sqldomain normalization)Test plan
compute_stats.pyoutputs 737 runs matching all original tex numbers🤖 Generated with Claude Code