2026
Breaking the Reversal Curse in Autoregressive Language Models via Identity Bridge
ICML 2026spotlight
Autoregressive large language models (LLMs) have achieved remarkable success in many complex tasks, yet they can still fail in very simple logical reasoning such as the "reversal curse" --- when trained on forward knowledge data of the form "$A \rightarrow B$" (e.g., Alice's husband is Bob), the mod…