Skip to content

Agent guidance contains verbatim aichat-matrix prompts — the eval is measuring recall of planted examples, not generalization #1763

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
malibio merged 7 commits into from
Jul 28, 2026
Merged

Agent guidance contains verbatim aichat-matrix prompts — the eval is measuring recall of planted examples, not generalization #1763

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
370782e
Decontaminate agent guidance of eval prompts and bind node_type to RE…
malibio Jul 28, 2026
c4e4e85
Address review: widen contamination guard to all model-facing guidance
malibio Jul 28, 2026
da12629
Close remaining guard gaps from re-review
malibio Jul 28, 2026
805a60e
Record the honest post-decontamination baseline (9/12)
malibio Jul 28, 2026
46bfed4
Record control-experiment results and the scenario-3 false pass in th…
malibio Jul 28, 2026
4755c9b
Close the remaining review suggestions on the baseline artifact
malibio Jul 28, 2026
23612bc
Correct the 8c characterization and hedge the generalization claim
malibio Jul 28, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Loading