@inproceedings{zhu-etal-2026-memorization,
    title = "Memorization, Emergence, and Explaining Reversal Failures: A Controlled Study of Relational Semantics in {LLM}s",
    author = "Zhu, Yihua  and
      Liu, Qianying  and
      Wang, Jiaxin  and
      Cheng, Fei  and
      Liu, Chaoran  and
      Aizawa, Akiko  and
      Kurohashi, Sadao  and
      Shimodaira, Hidetoshi",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/check-for-anonymous-pdfs/2026.acl-long.1688/",
    pages = "36425--36443",
    ISBN = "979-8-89176-390-6",
    abstract = "Autoregressive LLMs perform well on relational tasks that require linking entities via relational words (e.g., father/son, friend), but it is unclear whether they learn the logical semantics of such relations (e.g., symmetry and inversion logic) and, if so, whether reversal-type failures arise from missing relational semantics or left-to-right order bias. We propose a controlled Knowledge Graph-based synthetic framework that generates text from symmetric/inverse triples, train GPT-style autoregressive models from scratch, and evaluate memorization, logical inference, and in-context generalization to unseen entities to address these questions. We find a sharp phase transition in which relational semantics emerge with sufficient logic-bearing supervision, even in shallow (2{--}3 layer) models, and that successful generalization aligns with stable intermediate-layer signals. Finally, order-matched forward/reverse tests and a diffusion baseline indicate that reversal failures are primarily driven by autoregressive order bias rather than deficient inversion semantics."
}