diff options
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 37 |
1 files changed, 18 insertions, 19 deletions
@@ -36,28 +36,27 @@ mathematics. ## Prompt fidelity -The historical two-call Prompt-A/Prompt-B, surface-renaming prompts, and -review/repair prompts recovered from `PutnamVariants@c3bed737` remain byte +The consolidated Prompt-A/Prompt-B, surface-renaming prompts, and +review/repair prompts from `PutnamVariants@c3bed737` remain byte pinned in `src/gap_pipeline/prompts.py`, `PROMPT_SHA256SUMS`, and -`tests/test_prompts.py`. They document the actual original generator. +`tests/test_prompts.py`. -The executable manuscript-aligned prompts are in -`src/gap_pipeline/kernel_prompts.py`. They refine the historical intent into -five explicit contracts because the recovered original generator only made -two calls and did not emit a concrete DAG, applied replacement map, or -node-by-node diffusion trace. This distinction is deliberate and auditable, -not hidden as prompt identity. +The default executable path uses the expanded stage prompts in +`src/gap_pipeline/kernel_prompts.py` to execute the five operations specified +in Section 3.2 and Appendix D.2 as separately auditable calls. The five judges +use the byte-pinned Appendix F.3 system and user templates verbatim. The +consolidated and expanded interfaces are both retained in source; the expanded +five-stage path is the reproduction entry point. The OpenAI adapter does not send `temperature`; this is compatible with `o3`, whose supported value is its default. See `STAGE_MAP.md` for the exact -paper-to-code and historical-source maps. +paper-to-code map. `ProofDAG` supports branching dependencies and validates topological order, known dependencies, acyclicity, and terminal connectivity. Replacement plans record the exact old/new value, source node, guard condition, and guard justification. Diffusion must preserve every node ID, dependency, and method -label. Judges must cover every proof-node ID and replacement-slot ID before -their verdict counts. +label. Judges must cover every proof-node ID before their verdict counts. ## Install and test @@ -123,13 +122,13 @@ PYTHONPATH=src python -m gap_pipeline.cli export-release \ Kernel verification uses `J=5` judges, requires `K=2` consecutive unanimous rounds for the same complete provenance bundle, and allows at most `T=15` -rounds. A rejected round resets the streak and reruns stages 3--5 from a new -guarded replacement plan using the judge feedback. Every call, stage output, -iteration, and final record is saved under the chosen run directory. +rounds. A rejected round resets the streak and applies the judge feedback to +the previous bundle, rerunning stages 3--5 with an explicit minimal-repair +instruction. Every call, stage output, iteration, and final record is saved +under the chosen run directory. ## Scope -This package demonstrates and tests the one-item software path. It does not -reconstruct unavailable proposal/rejection logs from the original 1,051-item -generation run, and a successful LLM verification loop is not a substitute for -the separate blinded mathematical audit described in the rebuttal plan. +This package demonstrates and tests the complete one-item software path. The +same CLI and runner can be applied over additional item IDs to generate a +larger release. |