#natural language process...
Sep 2026
Constrained Decoding Eliminates Structural Failures in Small LLMs but Reveals a Scale-Dependent Semantic Gap
It is found that CD eliminates all structural failures across all models, but content accuracy reveals a persistent semantic gap that is scale-dependent: type coercion failures are fully CD-rescuable, while instruction-semantic failures remain CD-resistant.
Akash Chavan
· arXiv.org · 2 citations