← Complete research archive
Language compilerResult66 lines

R12 Complete-Compiler Parameter-Islands Result

Protocol: r12 referential literal pointer compiler v1 3 islands development Decision: PARTIAL MECHANISM WIN; FAIL FROZEN PROMOTION GATE

R12_REFERENTIAL_LITERAL_POINTER_COMPILER_ISLANDS_RESULT.mdOpen original Markdown ↗

R12 Complete-Compiler Parameter-Islands Result

Protocol: r12_referential_literal_pointer_compiler_v1_3_islands_development Decision: PARTIAL MECHANISM WIN; FAIL FROZEN PROMOTION GATE

Run and custody

Job 692992 completed on a verified H100 PCIe on evc28 with exit code zero in 10m10s. It used the frozen one-epoch / 1,514-update v1.1 training schedule, seed 2026071805, 8,658,701 trainable compiler parameters, and 133,740,365 total parameters. Shohin remained frozen.

ItemSHA-256
initial adapter state1e9301b55eecf1b1be4598a2e0e3a25bb09d989c2a8d7f37fc09dc310d599abc
final adapter state93d3aef45e818c61f9bb7857df45cb3086ba1b7e2b4d6478049913f3dc0516b3
adapter file63f735d7fa275b8347e25d3645914fc24d08e24b0c412f27c292a561ebb65af2
development result873254e5dde203ea89f2b55b68948ec64f13a3e0536e9d38281f88ad0539205d
Slurm logd017fbd983100d37303f5dfb5d05d9d84693501644379b114af43f4cb8564d99

The adapter, result, and log are hash-matched between Newton and the Mac. Confirmation remained absent from Newton and access is zero.

Frozen gates

MetricGateResultPass
operation-kind accuracy>=90%61.328%no
initial-state joint exact>=45%45.996%yes
answer accuracy>=45%43.311%no
semantic-program exact>=30%23.438%no
full pointer exact>=10%10.645%yes
paraphrase answer>=45%59.766%yes
canonical answer>=35%32.422%no
canonical + paraphrase both pointer-exact>=1/5120/512no

Three of eight gates pass. The arm is not promoted and v1.1 confirmation remains sealed.

Surface result

SurfaceKindInitial jointProgram exactAnswer
canonical46.289%45.508%15.820%32.422%
order twin53.711%44.336%19.336%41.016%
binding twin46.289%44.922%15.820%40.039%
paraphrase99.023%49.219%42.773%59.766%

The separate semantic island fixes v1.2's optimization collapse: training kind loss reaches zero, and exposed-development answer accuracy rises from 18.994% to 43.311%. It also exceeds the v1.1 answer result by 13.916 percentage points and lifts initial-state joint exact by 27.148 points.

The failure is now lexical and surface-specific. The island transfers almost perfectly to the unseen front/rear paraphrase renderer but not to the unseen earlier/later canonical renderer. Two fixed training renderer families do not identify renderer-invariant lexical semantics, even with 96,000 examples.

Scientific consequence

Parameter islands are supported as an optimization mechanism: structural and semantic paths can learn concurrently without the v1.2 chance-classifier collapse. They are not yet a complete compiler result. Repeating seeds or tuning against this exposed development split is unauthorized.

The next experiment must change the educational support, not the H100 budget: train on many factorized combinations of intro frames, argument orders, direction lexemes, operation ordinals, distractor placements, and query frames; evaluate unseen combinations separately from truly unseen lexemes; and compare v1.3 against favorable ordinary parser and oracle controls.