Papers
arxiv:2501.13491

RECALL: Library-Like Behavior In Language Models is Enhanced by Self-Referencing Causal Cycles

Published on Jan 23, 2025
Authors:
,
,
,
,
,
,
,
,

Abstract

Large language models suffer from unidirectional causality limitations known as the reversal curse, but a self-referencing causal cycle mechanism enables contextual recall despite this constraint.

We introduce the concept of the self-referencing causal cycle (abbreviated RECALL) - a mechanism that enables large language models (LLMs) to bypass the limitations of unidirectional causality, which underlies a phenomenon known as the reversal curse. When an LLM is prompted with sequential data, it often fails to recall preceding context. For example, when we ask an LLM to recall the line preceding "O say does that star-spangled banner yet wave" in the U.S. National Anthem, it often fails to correctly return "Gave proof through the night that our flag was still there" - this is due to the reversal curse. It occurs because language models such as ChatGPT and Llama generate text based on preceding tokens, requiring facts to be learned and reproduced in a consistent token order. While the reversal curse is often viewed as a limitation, we offer evidence of an alternative view: it is not always an obstacle in practice. We find that RECALL is driven by what we designate as cycle tokens - sequences that connect different parts of the training data, enabling recall of preceding tokens from succeeding ones. Through rigorous probabilistic formalization and controlled experiments, we demonstrate how the cycles they induce influence a model's ability to reproduce information. To facilitate reproducibility, we provide our code and experimental details at https://anonymous.4open.science/r/remember-B0B8/.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2501.13491
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2501.13491 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2501.13491 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2501.13491 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.