feat: evidence-bound context-position and provenance ablations - #907
Merged
Conversation
tcconnally
force-pushed
the
feat/issue-904-context-position-ablation
branch
from
August 3, 2026 13:12
c9b1c11 to
e02d26c
Compare
- Add offline context-position ablation harness (beginning/middle/end/shuffled/provenance_ranked) - Three context conditions: full, resolver-selected, resolver-selected+contract-anchor - Deterministic resolver filters quarantined/untrusted/superseded evidence - Hard gates: exact correctness, OOD abstention, poisoned-evidence rejection, attention-is-diagnostic-only negative control - Signed hash-bound report (signature_sha256) suitable for claims.json - Frozen hand-authored corpus with provenance metadata Closes #904
tcconnally
force-pushed
the
feat/issue-904-context-position-ablation
branch
from
August 3, 2026 13:12
e02d26c to
cc8f900
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Verification
python benchmark/context_position/run.py: correctness=1.0, all 4 gates PASS, exit 0pytest tests/test_context_position_ablation.py: 12 passedpytest tests/test_context_position_ablation.py tests/test_selection_eval.py: 16 passedCloses #904