Document main JSONLite continuation and matched evaluation

Documentation-only update. Base weights, tokenizer, native runner and source bundle are unchanged.

PACKAGING_VERIFICATION.json

@@ -18,5 +18,5 @@
     "scope": "Packaging smoke only. A plausible but repetitive continuation is not an accuracy measurement.",
     "matches_historical_prefix": true
   },
-  "documentation_update": "Native JSON descendant SFT-100 documented; pretrained base weights/tokenizer/runner/source bundle unchanged."
+  "documentation_update": "Main JSONLite descendant includes additional cardinality/replay distillation SFT; base weights, tokenizer, runner and bundle unchanged."
 }

README.md

@@ -176,14 +176,19 @@ not establish blanket underlying-content redistribution clearance.
 ## Native JSON descendant update
 
 The current [OpenSML-JSONLite-175M-Instruct](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M)
-continues this pretrained ancestor through grounded QA, exact field extraction,
-focused extraction and a final 15,000-record native JSON SFT run (469 updates).
-Its model generates the entire JSON object from source text and a supplied
-schema in one request, without a fieldwise assembler, repair or constrained
-decoding path. Native JSON held-out correctness is 1,191/1,444 (82.48%), with
-91.31% field accuracy and 90.48% extracted-value F1; the 54-case local diagnostic
-scores 42/54. These belong to the post-trained descendant, not this base model.
-Scope, artifacts, source revisions and remaining errors are in the descendant's
-[technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md).
-The base weights, base tokenizer, raw-completion runner and base evaluation
-results are unchanged.
+continues this pretrained base through grounded QA, requested-field and focused extraction,
+a 15,000-presentation native JSON foundation, and an additional focused extraction SFT.
+The added stage combines original JSON TRAIN replay, matched zero/one/two/three-value
+examples, and training-only frozen-parent distillation. The main model generates the
+entire JSON object itself from source text and a supplied schema, with no assembly or repair.
+
+On the same previously observed 1,444-request native JSON evaluation, complete-request
+correctness is **1,199/1,444 (83.03%)**, compared with 1,191/1,444 for the prior native release.
+Field accuracy is 91.36%, value F1 90.40%, missing-field accuracy 92.79%, and schema validity
+99.79%. The canonical subset is 1,163/1,388 (83.79%); observed playground correctness is
+44/54. These are descendant results, not scores of this pretrained base. Some domains and
+value precision/recall regress slightly, and strict automatic retention gates were not met.
+See the [JSONLite technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md)
+for full training provenance, matched comparison and limitations.
+
+The base weights, tokenizer, inference runner and source bundle are unchanged by this documentation update.

RUN_GUIDE.md

@@ -61,14 +61,19 @@ following or extraction. See the model card and report for measured scope.
 ## Native JSON descendant update
 
 The current [OpenSML-JSONLite-175M-Instruct](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M)
-continues this pretrained ancestor through grounded QA, exact field extraction,
-focused extraction and a final 15,000-record native JSON SFT run (469 updates).
-Its model generates the entire JSON object from source text and a supplied
-schema in one request, without a fieldwise assembler, repair or constrained
-decoding path. Native JSON held-out correctness is 1,191/1,444 (82.48%), with
-91.31% field accuracy and 90.48% extracted-value F1; the 54-case local diagnostic
-scores 42/54. These belong to the post-trained descendant, not this base model.
-Scope, artifacts, source revisions and remaining errors are in the descendant's
-[technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md).
-The base weights, base tokenizer, raw-completion runner and base evaluation
-results are unchanged.
+continues this pretrained base through grounded QA, requested-field and focused extraction,
+a 15,000-presentation native JSON foundation, and an additional focused extraction SFT.
+The added stage combines original JSON TRAIN replay, matched zero/one/two/three-value
+examples, and training-only frozen-parent distillation. The main model generates the
+entire JSON object itself from source text and a supplied schema, with no assembly or repair.
+
+On the same previously observed 1,444-request native JSON evaluation, complete-request
+correctness is **1,199/1,444 (83.03%)**, compared with 1,191/1,444 for the prior native release.
+Field accuracy is 91.36%, value F1 90.40%, missing-field accuracy 92.79%, and schema validity
+99.79%. The canonical subset is 1,163/1,388 (83.79%); observed playground correctness is
+44/54. These are descendant results, not scores of this pretrained base. Some domains and
+value precision/recall regress slightly, and strict automatic retention gates were not met.
+See the [JSONLite technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md)
+for full training provenance, matched comparison and limitations.
+
+The base weights, tokenizer, inference runner and source bundle are unchanged by this documentation update.

SHA256SUMS

@@ -1,12 +1,12 @@
 59ee3f5b0b4e357e34a390a72febcc13eac5493fb4a4c2d7ecdf6ba5be324ff2  CITATION.cff
 9c6d4c439acf90f9187ba6cf11f070afb3ea1515d488b52c381adf9356a1d613  EVALUATION_RESULTS.json
 c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  LICENSE
-bdbb7150a49d2a0303709974aba2a1337077f4f855f6ea21ffa979cf3a3b66a9  PACKAGING_VERIFICATION.json
+27478e40538a7777e0e1f8aed83f16afa6d9cf47084f6bfd39c7f8a9df515fb9  PACKAGING_VERIFICATION.json
 14819fc05509546bb8218153cf8494c6a3c4e2e4744ce504bdc21ef0d7d1ed59  PRETRAINING_SAMPLES.json
-aa65acf8e5bd2e23b97e32a7c55ef6145dc60ecb7583908b13842a9e48d4c37c  README.md
-023cf9b387b8936ad6cf66fd3a616bd2b20af1e80442b8936a24dc8c50992791  RUN_GUIDE.md
+f97034ba3ecbbb05f200a48cb9813ea5eb9c9e46c19f04496db14b03251a3967  README.md
+52863aa0d0b0fc25cad46c07830c244b9fa576f38b29fe39d2d086bbf8e78fae  RUN_GUIDE.md
 c3bd058575a0313f716dabb8de8a3d7b729ff413dae5f74b06525c0737dcfac3  SOURCE_FILES.json
-3a7021f7e242496e3121f8e4736ce56c8bd8e62d665d595191faadcb01246080  TECHNICAL_REPORT.md
+3ff056103137004972aaa69e1f50783b70534d951a905579cb72f151a4170d44  TECHNICAL_REPORT.md
 3235c99f09bfab6e5ee95e64a382a4975978ec7f8988d33a71d4aaeb0215bf15  THIRD_PARTY_NOTICES.md
 15a0e3b511b97503a611c0da838a7fde1c06ebed34c7103f7cda1782cc6378a4  checkpoint_metadata.json
 22eaa19ba2346558711ff5245b4a1e581a431b7b1c5df3c2342f52407ebbc74a  checkpoint_provenance.json

TECHNICAL_REPORT.md

@@ -154,14 +154,19 @@ underlying-source redistribution clearance.
 ## Native JSON descendant update
 
 The current [OpenSML-JSONLite-175M-Instruct](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M)
-continues this pretrained ancestor through grounded QA, exact field extraction,
-focused extraction and a final 15,000-record native JSON SFT run (469 updates).
-Its model generates the entire JSON object from source text and a supplied
-schema in one request, without a fieldwise assembler, repair or constrained
-decoding path. Native JSON held-out correctness is 1,191/1,444 (82.48%), with
-91.31% field accuracy and 90.48% extracted-value F1; the 54-case local diagnostic
-scores 42/54. These belong to the post-trained descendant, not this base model.
-Scope, artifacts, source revisions and remaining errors are in the descendant's
-[technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md).
-The base weights, base tokenizer, raw-completion runner and base evaluation
-results are unchanged.
+continues this pretrained base through grounded QA, requested-field and focused extraction,
+a 15,000-presentation native JSON foundation, and an additional focused extraction SFT.
+The added stage combines original JSON TRAIN replay, matched zero/one/two/three-value
+examples, and training-only frozen-parent distillation. The main model generates the
+entire JSON object itself from source text and a supplied schema, with no assembly or repair.
+
+On the same previously observed 1,444-request native JSON evaluation, complete-request
+correctness is **1,199/1,444 (83.03%)**, compared with 1,191/1,444 for the prior native release.
+Field accuracy is 91.36%, value F1 90.40%, missing-field accuracy 92.79%, and schema validity
+99.79%. The canonical subset is 1,163/1,388 (83.79%); observed playground correctness is
+44/54. These are descendant results, not scores of this pretrained base. Some domains and
+value precision/recall regress slightly, and strict automatic retention gates were not met.
+See the [JSONLite technical report](https://huggingface.co/OpenSML/OpenSML-JSONLite-175M/blob/main/TECHNICAL_REPORT.md)
+for full training provenance, matched comparison and limitations.
+
+The base weights, tokenizer, inference runner and source bundle are unchanged by this documentation update.