BWhat changed
Released a documented English/Korean instruction-tuned language-model resource with bilingual tokenization, pretraining and alignment details and public/in-house evaluation.
Scientific scope
Released a documented English/Korean instruction-tuned language-model resource with bilingual tokenization, pretraining and alignment details and public/in-house evaluation. A model-resource contribution, not a new invention of Transformer, RoPE or GQA. The original released checkpoint has research/non-commercial restrictions; open weights must not be described as an unrestricted permissive license. Claims of general expert competence are not adopted.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. A model-resource contribution, not a new invention of Transformer, RoPE or GQA. The original released checkpoint has research/non-commercial restrictions; open weights must not be described as an unrestricted permissive license. Claims of general expert competence are not adopted.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- EXAONE 3.0 7.8B Instruction Tuned Language Model
- LG AI Research: EXAONE bilingual instruction-model resource — author/company record
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction