Small LLMs perform no-CoT two-hop reasoning on synthetic facts with real bridges
GabrielKS
Abstract
TL;DR A good understanding of where LLM no-chain of thought (no-CoT) two-hop reasoning succeeds and fails is important for monitorability as well as for designing belief instillation-based misalignment evaluations. Prior work suggests that LLMs are limited in their ability to perform no-CoT two-hop reasoning with pairs of fabricated, fine-tuned propositions. Here, I show on Qwen3.6-35B-A3B and Qwen3-8B that this limitation only holds if the bridge entity connecting the propositions is also fabricated or particularly obscure — models successfully compose such fabricated propositions if the bridge entity is real, and success rates vary with how well-known it is. I also demonstrate that, with real and well-known bridge entities, the source of the other entities barely matters — models achieve roughly comparable two-hop reasoning accuracies whether each of the remaining entities is real or synthetic. Results hold with both SFT and synthetic document fine-tuning (SDF). These results show a previously undemonstrated latent reasoning capability of small LLMs and provide a controlled setting for more mechanistic analyses of two-hop reasoning. Code: https://github.com/redwoodresearch/sdf-tw