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

• URL: http://arxiv.org/abs/2501.13491v1
• Date: Thu, 23 Jan 2025 09:14:07 GMT
• Title: RECALL: Library-Like Behavior In Language Models is Enhanced by Self-Referencing Causal Cycles
• Authors: Munachiso Nwadike, Zangir Iklassov, Toluwani Aremu, Tatsuya Hiraoka, Velibor Bojkovic, Benjamin Heinzerling, Hilal Alqaubeh, Martin Takáč, Kentaro Inui,
• Abstract summary: We introduce the concept of the self-referencing causal cycle (abbreviated RECALL)<n>It enables large language models to bypass the limitations of unidirectional causality.<n>We find that RECALL is driven by what we designate as cycle tokens.
• Score: 18.140067201462884
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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/.

Related papers

• Positional Fragility in LLMs: How Offset Effects Reshape Our Understanding of Memorization Risks [22.636913794637437]
  We show that verbatim memorization is most strongly triggered by short prefixes drawn from the beginning of the context window.<n>We show that shifting sensitive data deeper into the context window suppresses both extractable memorization and degeneration.
  arXiv  Detail & Related papers  (2025-05-19T14:28:35Z)
• Understanding the Repeat Curse in Large Language Models from a Feature Perspective [10.413608338398785]
  Large language models (LLMs) often suffer from repetitive text generation.<n>We propose a novel approach, "Duplicatus Charm", to induce and analyze the Repeat Curse.
  arXiv  Detail & Related papers  (2025-04-19T07:53:37Z)
• Is the Reversal Curse a Binding Problem? Uncovering Limitations of Transformers from a Basic Generalization Failure [14.07889703663922]
  LLMs exhibit a basic generalization failure known as the Reversal Curse. We conjecture that the Reversal Curse in LLMs is a manifestation of the long-standing binding problem in cognitive science, neuroscience and AI.
  arXiv  Detail & Related papers  (2025-04-02T17:38:03Z)
• Interpreting the Repeated Token Phenomenon in Large Language Models [31.1226642501095]
  Large Language Models (LLMs) often fail to accurately repeat a single word when prompted to, and instead output unrelated text. We aim to explain the causes for this phenomenon and link it to the concept of attention sinks'' Our investigation identifies the neural circuit responsible for attention sinks and shows how long repetitions disrupt this circuit.
  arXiv  Detail & Related papers  (2025-03-11T21:40:58Z)
• From Loops to Oops: Fallback Behaviors of Language Models Under Uncertainty [67.81977289444677]
  Large language models (LLMs) often exhibit undesirable behaviors, such as hallucinations and sequence repetitions. We categorize fallback behaviors - sequence repetitions, degenerate text, and hallucinations - and extensively analyze them. Our experiments reveal a clear and consistent ordering of fallback behaviors, across all these axes.
  arXiv  Detail & Related papers  (2024-07-08T16:13:42Z)
• The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse and More [27.731438642876114]
  We study the reversal curse, where models cannot recall information when probed in a different order than was encountered during training. We find that the factorization curse is an inherent failure of the next-token prediction objective used in popular large language models. Our results uncover a promising path forward: factorization-agnostic objectives can significantly mitigate the reversal curse and hint at improved knowledge storage and planning capabilities.
  arXiv  Detail & Related papers  (2024-06-07T18:00:37Z)
• Mitigating Reversal Curse in Large Language Models via Semantic-aware Permutation Training [57.771940716189114]
  We show that large language models (LLMs) suffer from the "reversal curse" The root cause of the reversal curse lies in the different word order between the training and inference stage. We propose Semantic-aware Permutation Training (SPT) to address this issue.
  arXiv  Detail & Related papers  (2024-03-01T18:55:20Z)
• OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation [124.9008419182485]
  We present OPERA, a novel MLLM decoding method grounded in an Over-trust Penalty and a Retrospection-Allocation strategy. Our approach begins with an interesting observation that, most hallucinations are closely tied to the knowledge aggregation patterns in the self-attention matrix. Based on the observation, OPERA introduces a penalty term on the model logits during the beam-search decoding to mitigate the over-trust issue.
  arXiv  Detail & Related papers  (2023-11-29T18:57:07Z)
• An Analysis and Mitigation of the Reversal Curse [70.13419502543915]
  Recent research observed a noteworthy phenomenon in large language models (LLMs) The reversal curse is that when dealing with two entities, $a$ and $b$, LLMs excel in handling sequences in the form of $aRb$,'' but encounter challenges when processing $bR-1a$''
  arXiv  Detail & Related papers  (2023-11-13T17:01:12Z)
• Explaining Emergent In-Context Learning as Kernel Regression [61.57151500616111]
  Large language models (LLMs) have initiated a paradigm shift in transfer learning. In this paper, we investigate the reason why a transformer-based language model can accomplish in-context learning after pre-training. We find that during ICL, the attention and hidden features in LLMs match the behaviors of a kernel regression.
  arXiv  Detail & Related papers  (2023-05-22T06:45:02Z)
• Back to the Future: Unsupervised Backprop-based Decoding for Counterfactual and Abductive Commonsense Reasoning [79.48769764508006]
  generative language models (LMs) can be trained to condition only on the past context or to perform narrowly scoped text-infilling. We propose DeLorean, a new unsupervised decoding algorithm that can flexibly incorporate both the past and future contexts. We demonstrate that our approach is general and applicable to two nonmonotonic reasoning tasks: abductive text generation and counterfactual story revision.
  arXiv  Detail & Related papers  (2020-10-12T17:58:43Z)
• Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgetting [100.75479161884935]
  We propose a novel training paradigm called Remembering for the Right Reasons (RRR) RRR stores visual model explanations for each example in the buffer and ensures the model has "the right reasons" for its predictions. We demonstrate how RRR can be easily added to any memory or regularization-based approach and results in reduced forgetting.
  arXiv  Detail & Related papers  (2020-10-04T10:05:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.