paper-with-me

홈 › Papers

An Analysis and Mitigation of the Reversal Curse

2023-11-13 · Ang Lv, Kaiyi Zhang, Shufang Xie, Quan Tu, Yuhan Chen, Ji-Rong Wen, Rui Yan

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.

📄 PDF Abstract BibTeX arXiv:2311.07468

Code (1)

trestad/mitigating-reversal-curse 공식 구현 pytorch

Tasks

DenoisingLanguage Modelling

Methods 이 논문이 사용한 방법론

Focus 설명 없음
GLM GLM is a bilingual (English and Chinese) pre-trained transformer-based language model that follow the traditional architecture of decoder-only autoregressive language…

Similar Papers 제목 키워드 기반

Towards a Theoretical Understanding of the 'Reversal Curse' via Training Dynamics

2024-05-07 · Hanlin Zhu, Baihe Huang, Shaolun Zhang, Michael Jordan 외

Auto-regressive large language models (LLMs) show impressive capacities to solve many complex reasoning tasks while struggling with some simple logical reasoning tasks such as inverse search: when trained on '$A \to B$' …

Logical Reasoning

DiffER: Diffusion Entity-Relation Modeling for Reversal Curse in Diffusion Large Language Models

2026-01-12 · Shaokai He, Kaiwen Wei, Xinyi Zeng, Xiang Chen 외 arxiv

The "reversal curse" refers to the phenomenon where large language models (LLMs) exhibit predominantly unidirectional behavior when processing logically bidirectional relationships. Prior work attributed this to autoregr…

A Theoretical Analysis of Why Masked Diffusion Models Mitigate the Reversal Curse

2026-02-02 · Moongyu Jeon, Sangwoo Shin, BumJun Kim, Kyelim Lee 외 arxiv

Autoregressive language models (ARMs) suffer from the reversal curse: after learning ''$A$ is $B$,'' they often fail on the reverse query ''$B$ is $A$.'' Masked diffusion language models (MDMs) exhibit this failure in a …

The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse and More

2024-06-07 · Ouail Kitouni, Niklas Nolte, Diane Bouchacourt, Adina Williams 외

Today's best language models still struggle with hallucinations: factually incorrect generations, which impede their ability to reliably retrieve information seen during training. The reversal curse, where models cannot …

Information RetrievalRetrieval

Is the Reversal Curse a Binding Problem? Uncovering Limitations of Transformers from a Basic Generalization Failure

2025-04-02 · Boshi Wang, Huan Sun

Despite their impressive capabilities, LLMs exhibit a basic generalization failure known as the Reversal Curse, where they struggle to learn reversible factual associations. Understanding why this occurs could help ident…

Arithmetic ReasoningData Augmentation