Pronunciation cache limited to the vocabulary words; README wording

README.md

@@ -31,7 +31,7 @@ writes no คำสร้อย. One sample is drawn per topic at temperature 0.
 | `model/config.json`, `TRAINED.json`, `train.log` | model configuration, training record and log |
 | `lexicon.json` | the vocabulary: 62,856 words with their pronounced syllables |
 | `allow.json` | allowlist of the decoder's stutter ban |
-| `g2p_raw_cache.json` | pronunciation cache (word -> phonemes) |
+| `g2p_raw_cache.json` | pronunciation cache for the 62,856 vocabulary words (word -> phonemes) |
 | `config.json`, `configuration_klong4.py`, `modeling_klong4.py` | the `transformers` wrapper |
 | `src/klong4/` | the project's code, used by the wrapper |
 | `SHA256SUMS` | checksums of the weight and data files above |
@@ -74,21 +74,23 @@ Notes:
 
 ## Evaluation (from the paper)
 
-On 75 held-out topics, one sample per topic, judged by three independent runs of an LLM judge (Claude Opus 5.5) in the
-same pools as the human stanzas of those topics:
+On 75 held-out stanzas and their topics (63 distinct), one sample per item, judged by three independent runs of an LLM
+judge (Claude Opus 5.5) in the same pools as the human stanzas:
 
 | | Model | Human verse |
 |---|---|---|
-| Passes form and all three rhymes | 100% | 100% (selected) |
+| Passes the form checker, three rhymes included | 100% | 100% (selected) |
 | Lexically clean | 66.7% | 92.0% |
 | On topic (alignment >= 4 of 5) | 72.0% | 88.0% |
 | Clean and on topic | 50.7% | 81.3% |
 
-The difference in the last row is -30.7 points (95% CI [-44.0, -17.3]). No generated line is copied from the corpora.
+The difference in the last row is -30.7 points (95% CI [-44.0, -17.3]). No generated line equals a line of the corpora searched, and
+none shares more than five consecutive syllables with them.
 
 ## Limitations
 
-- Every stanza satisfies the form, but a third contain a line the judge considers ill-formed, and the vocabulary is
+- Every stanza passes the automatic form checker, which counts syllables as the pronunciation converter reads them
+  (it misreads some words), but a third contain a line the judge considers ill-formed, and the vocabulary is
   narrower and more repetitive than human verse (6.8% rare words against 13.9%).
 - The training set is small (2,001 stanzas).
 - The numbers are judgments of one LLM judge family, not of expert readers.

SHA256SUMS

@@ -1,4 +1,4 @@
 f053475b9be734873050775abc9fcd51cc01f271ce80d22352126ed9e97e8916  model/model.pt
 5e47c92e00fd240544a4acf8fe8f95fd30a1b79a7899d08023039ea810e3c228  lexicon.json
 fdcb862c4dcab8f370dc0bac46d9066ddee92fd3728a4ffdf7d12b99d9e0eaf5  allow.json
-7177cfba9e087512f0c7e805440ad64533b69f34ad7c8a9803d97d84c8be92b3  g2p_raw_cache.json
+8d4b4ef7c9860fd4181c1eb9cd38856a900b7b5899698f7ff95c01c35c774466  g2p_raw_cache.json