An Analysis and Mitigation of the Reversal Curse

TL;DR

This paper investigates the reversal curse in large language models, a phenomenon where models perform well on forward relations but struggle with their inverses, revealing limitations linked to training objectives.

Contribution

First comprehensive analysis of the reversal curse in LLMs, linking it to training objectives and highlighting a key limitation in current models.

Findings

Reversal curse stems from next-token prediction training

Models excel in forward relation tasks but struggle with inverse relations

Highlights need for improved training strategies to address this limitation

Abstract

Recent research observed a noteworthy phenomenon in large language models (LLMs), referred to as the ``reversal curse.'' The reversal curse is that when dealing with two entities, denoted as $a$ and $b$, connected by their relation $R$ and its inverse $R^{-1}$, LLMs excel in handling sequences in the form of ``$aRb$,'' but encounter challenges when processing ``$bR^{-1}a$,'' whether in generation or comprehension. For instance, GPT-4 can accurately respond to the query ``Tom Cruise's mother is?'' with ``Mary Lee Pfeiffer,'' but it struggles to provide a satisfactory answer when asked ``Mary Lee Pfeiffer's son is?'' In this paper, we undertake the first-ever study of how the reversal curse happens in LLMs. Our investigations reveal that the reversal curse can stem from the specific training objectives, which become particularly evident in the widespread use of next-token prediction within most causal language models. We hope this initial investigation can draw more attention to the reversal curse, as well as other underlying limitations in current LLMs.