Repository navigation
Conditional κ = 0.001662039077758742 — U10 (two new triangle-hybrid slots) on PR942's T37 owners - #987
Merged
Conversation
rohanarun
force-pushed
the
u10-t21-slots
branch
from
October 13, 2026 16:13
36b76ed to
fa7b789
Compare
…ditional kappa 0.001662039077758742
Adds research/complex-u10-t37: the U10 local word (PR869's U8 plus two new
triangle-hybrid slots found by a generalised slot search, X = 444 and 446 on the
entering pair 612/613), hosted by PR942's unchanged T37 owner map. Profile delta
{3:+4,4:-8,5:+4}; widths, prefixes and owners unchanged. Strict cold replay pending.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TGoMCHAqctZe4wTb4Qr7VV
rohanarun
force-pushed
the
u10-t21-slots
branch
from
October 13, 2026 16:13
fa7b789 to
0beb947
Compare
…receipts - prepare.py: rebind the inherited U8 provider identities in the T8 plan registry to the frozen U10 values (construction/u10/PROVIDER-U10.json); bind re-checks them on fresh locals. - final_binding.py: U10 profile binding reads the raw U10 word and both local checks. - evidence/strict: RESULT (PASS_U10_T37_STRICT_WITH_SOURCE_BOUND_ANCESTRY), FINAL-BINDING, STAGES, strict ASSEMBLY, payload controls, independent owner check, sub-stage lists. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TGoMCHAqctZe4wTb4Qr7VV
rohanarun
marked this pull request as ready for review
October 13, 2026 17:38
Closed
This was referenced Oct 13, 2026
This was referenced Oct 14, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Conditional κ = 0.001662039077758742: U10 (two new triangle-hybrid slots) on PR942's T37 owners
This is +7.8530044795×10⁻⁸ over the selected PR942 result (0.001661960547713947), about 28× PR942's own gain over PR890. The complex side binds; the bit supplier is PR589's, unchanged.
Strict cold replay: PASS.
python -B research/complex-u10-t37/verify.py --work /tmp/u10-t37 [--baseline799 /abs/completed-cold799] # PASS_U10_T37_STRICT_WITH_SOURCE_BOUND_ANCESTRY 0.001662039077758742The run is PR942's complete pipeline plus one
u10-nativestage. It covers:The cold PR799/PR741/PR589 ancestry was replayed fresh (
PASS_REBUILT_PR589_PR799_COMPOSITION). Receipts are inevidence/strict/(Linux, Python 3.13).What changes: only the local word (U8 → U10)
I generalised NewBArch's PR869 triangle-hybrid B delivery into an exhaustive slot search over the U8 word. The search covers:
It uses exact pre-read reconstruction and a chain-only frame re-optimiser, and every winner goes through the independent
audit_local.py. It finds two new legal slots outside block 0, on the entering pair 612/613 (X = 444 and 446):Each slot is
{3:+2, 4:−4, 5:+2}, like a PR869 slot, and the two are additive. U8{1:2548,2:338,3:446,4:162,5:72,8:8}becomes U10{1:2548,2:338,3:450,4:154,5:76,8:8}.Width 3778, width inventory, local prefixes (55/72 bits), A section, start/end frames, centres and every kept B frame are unchanged. PR942's T37 owners, order, edits, banks, 1024-bit prefix charge and outer recipe are therefore used byte-identically. The global profile is
333×U10 + {48:11086, 1:6}. PR942's all-replica bound, rerun on U10, has grid upper 0.001662039077758742, equal to the claim, so T37 stays grid-optimal for U10.The U10 word is an explicit recipe: 10 deletes, 10 inserts, 2 tails and 2 restores. NewBArch's unchanged
build_tri.pyapplies it to the pinned U8 word, giving content sha256776de990…. NewBArch's unchangedrun_controls.pyreports 14/14 outcomes:95 per-end layer mutations are rejected.
Package:
research/complex-u10-t37/It is PR942's
complex-t37-ownerswith these edits:verify.py: theu10-nativestage and U10 inputs.prepare.py:build_parentreads this package's CLAIM local profile;construction/u10/PROVIDER-U10.json. Bind re-checks every one against the fresh locals.final_binding.py: U10 checks and constants.all_t_bound.py: the U10 local profile.check_payload_controls.py: the new κ.construction/u10/(recipe, driver, pins, discovery code) andevidence/.Owner files and all inherited sources are byte-identical.
Credits:
build_tri.pyandrun_controls.py, used unchanged.🤖 Generated with Claude Code
https://claude.ai/code/session_01TGoMCHAqctZe4wTb4Qr7VV