D-RPC compresses reasoning into a dynamic bank of reusable paths to produce consistent teacher rationales, outperforming standard distillation baselines on five reasoning benchmarks while using fewer tokens.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Structural Rationale Distillation via Reasoning Space Compression
D-RPC compresses reasoning into a dynamic bank of reusable paths to produce consistent teacher rationales, outperforming standard distillation baselines on five reasoning benchmarks while using fewer tokens.