Preprint
Sep 2026
CPUNeSy: Controlling Model Writes for Reliable Neuro-Symbolic Reasoning
LLMs excel at recalling statistical patterns but degrade sharply when answers must be derived, especially on multi-hop chains. Delegating derivation to deterministic symbolic executors shifts reliability to whether model-generated premises are source-supported. We introduce CPUNeSy, a serving architecture that controls...
Ze-Yan Li, Si-Yuan Qiu, Shuai Zhao et al.
· 0 citations