Skip to content
arXiv cs.CL · Papers

Adapter Merging Reactivates Latent Reasoning Traces: A Mechanism Analysis

arXiv:2601.18350v5 Announce Type: replace Abstract: Large language models fine-tuned via a two-stage pipeline (domain adaptation followed by instruction alignment) can exhibit non-trivial interference after adapter merging, including the re-emergence of explicit reasoning traces under strict decoding. We study this phe