BWhat changed
Built and documented the original Qwen pretraining and human-alignment pipeline, including multilingual tokenization, context extension and chat models evaluated on tool use, code interpretation and specialized reasoning.
Scientific scope
Built and documented the original Qwen pretraining and human-alignment pipeline, including multilingual tokenization, context extension and chat models evaluated on tool use, code interpretation and specialized reasoning. One original family and training-system record. RoPE, SwiGLU, FlashAttention, RLHF and NTK-aware interpolation are inherited rather than attributed as Qwen inventions. Author benchmark results are not current product guarantees.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. One original family and training-system record. RoPE, SwiGLU, FlashAttention, RLHF and NTK-aware interpolation are inherited rather than attributed as Qwen inventions. Author benchmark results are not current product guarantees.
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.
- Qwen Technical Report
- Bai et al.: Qwen multilingual and tool-using language-model system — official research 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