CECoR decomposes multi-hop claims into steps, synthesizes training pairs via perturbation injection, and uses supervised fine-tuning plus reinforcement learning to improve factual error correction on multi-hop benchmarks.
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
-
Compositional Multi-hop Factual Error Correction via Decomposition-and-Injection
CECoR decomposes multi-hop claims into steps, synthesizes training pairs via perturbation injection, and uses supervised fine-tuning plus reinforcement learning to improve factual error correction on multi-hop benchmarks.