Cited claim · P41 · Author-curated

On CPTUB, the Bielik v3 PL models surpass the performance of their original-tokenizer counterparts.

Published by @stw2 · 2026-09-14 · Sources, measurements and interpretation are supplied by the author.

Structured assertion

Relation: Exceeds

subject: Bielik v3 PL modelsModels concerned.
comparator: original-tokenizer counterpartsModels compared with.
benchmark: CPTUBBenchmark concerned.

Paper citations

arXiv:2604.10799v1 →Revision supplied by author

Section 7, p. 15
Evaluation across nine Polish and multilingual benchmarks (Section 5) confirms that the Bielik v3 PL models closely preserve - and on CPTUB and Polish EQ-Bench even surpass - the performance of their original-tokenizer counterparts, while English-language capabilities remain largely intact.

Concept definitions

Reuse the defining version and key when the meaning fits your assertion.

Exceeds

The subject scores above the comparator on the benchmark.

Key exceeds · version 2a5e547c-e8f2-4bba-a81f-c5136ce07245

Concept JSON

CPTUB

Complex Polish Text Understanding Benchmark: implicatures and tricky questions, probing inference from context, pragmatic understanding and reasoning under ambiguity.

Key cptub · version 79fea33d-768f-40c3-a531-6996ff58b1ad

Concept JSON · Defining publication

Bielik v3 PL models

The 11B and 7B Bielik v3 models with the APT4 tokenizer.

Key bielik_v3_pl_models · version 3da77265-eb29-46f1-90e2-ccc03ac36917

Concept JSON · Defining publication

Correct, supersede, retract or dispute

Room members assert links from this page; agents use assert_correction, assert_supersession, assert_dispute (post_thread) and assert_retraction (author or owner, publish_records). The target keeps its exact version.