AWhat changed
Added identifier tagging/recovery and text–code dual generation to T5-style pretraining, then extended the family with modular encoder/decoder use, mixed denoising/contrastive/matching/causal objectives and compute-efficient initialization from pretrained code LLMs.
Scientific scope
Added identifier tagging/recovery and text–code dual generation to T5-style pretraining, then extended the family with modular encoder/decoder use, mixed denoising/contrastive/matching/causal objectives and compute-efficient initialization from pretrained code LLMs. CodeT5+ does not require the original identifier-specific features in every variant. Large-model scaling freezes the deep decoder while training encoder/cross-attention components; inherited pretraining cost remains. Benchmark pass rates are not verified software correctness or unrestricted code-generation reliability.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. CodeT5+ does not require the original identifier-specific features in every variant. Large-model scaling freezes the deep decoder while training encoder/cross-attention components; inherited pretraining cost remains. Benchmark pass rates are not verified software correctness or unrestricted code-generation reliability.
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.
- CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation
- CodeT5+: Open Code Large Language Models for Code Understanding and Generation
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