Glitch Tokens in Large Language Models: Categorization Taxonomy and Effective Detection

• URL: http://arxiv.org/abs/2404.09894v3
• Date: Fri, 19 Apr 2024 11:08:36 GMT
• Title: Glitch Tokens in Large Language Models: Categorization Taxonomy and Effective Detection
• Authors: Yuxi Li, Yi Liu, Gelei Deng, Ying Zhang, Wenjia Song, Ling Shi, Kailong Wang, Yuekang Li, Yang Liu, Haoyu Wang,
• Abstract summary: We introduce and explore the phenomenon of "glitch tokens", which are anomalous tokens produced by established tokenizers. We present categorizations of the identified glitch tokens and symptoms exhibited by LLMs when interacting with glitch tokens. We propose GlitchHunter, a novel iterative clustering-based technique, for efficient glitch token detection.
• Score: 23.382844182972807
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: With the expanding application of Large Language Models (LLMs) in various domains, it becomes imperative to comprehensively investigate their unforeseen behaviors and consequent outcomes. In this study, we introduce and systematically explore the phenomenon of "glitch tokens", which are anomalous tokens produced by established tokenizers and could potentially compromise the models' quality of response. Specifically, we experiment on seven top popular LLMs utilizing three distinct tokenizers and involving a totally of 182,517 tokens. We present categorizations of the identified glitch tokens and symptoms exhibited by LLMs when interacting with glitch tokens. Based on our observation that glitch tokens tend to cluster in the embedding space, we propose GlitchHunter, a novel iterative clustering-based technique, for efficient glitch token detection. The evaluation shows that our approach notably outperforms three baseline methods on eight open-source LLMs. To the best of our knowledge, we present the first comprehensive study on glitch tokens. Our new detection further provides valuable insights into mitigating tokenization-related errors in LLMs.

Related papers

• Learning on LLM Output Signatures for gray-box LLM Behavior Analysis [52.81120759532526]
  Large Language Models (LLMs) have achieved widespread adoption, yet our understanding of their behavior remains limited. We develop a transformer-based approach to process that theoretically guarantees approximation of existing techniques. Our approach achieves superior performance on hallucination and data contamination detection in gray-box settings.
  arXiv  Detail & Related papers  (2025-03-18T09:04:37Z)
• Attention Reallocation: Towards Zero-cost and Controllable Hallucination Mitigation of MLLMs [62.9348974370985]
  We propose attention reallocation (AttnReal) to mitigate hallucinations with nearly zero extra cost. Our approach is motivated by the key observations that, MLLM's unreasonable attention distribution causes features to be dominated by historical output tokens. Based on the observations, AttnReal recycles excessive attention from output tokens and reallocates it to visual tokens, which reduces MLLM's reliance on language priors.
  arXiv  Detail & Related papers  (2025-03-11T11:52:37Z)
• Demystifying Singular Defects in Large Language Models [61.98878352956125]
  In large language models (LLMs), the underlying causes of high-norm tokens remain largely unexplored. We provide both theoretical insights and empirical validation across a range of recent models. We showcase two practical applications of these findings: the improvement of quantization schemes and the design of LLM signatures.
  arXiv  Detail & Related papers  (2025-02-10T20:09:16Z)
• Forking Paths in Neural Text Generation [14.75166317633176]
  We develop a novel approach to representing uncertainty dynamics across individual tokens of text generation. We use our method to analyze LLM responses on 7 different tasks across 4 domains. We find many examples of forking tokens, including surprising ones such as punctuation marks.
  arXiv  Detail & Related papers  (2024-12-10T22:57:57Z)
• FIRP: Faster LLM inference via future intermediate representation prediction [54.897493351694195]
  FIRP generates multiple tokens instead of one at each decoding step. We conduct extensive experiments, showing a speedup ratio of 1.9x-3x in several models and datasets.
  arXiv  Detail & Related papers  (2024-10-27T15:53:49Z)
• Zero-Shot Detection of LLM-Generated Text using Token Cohesiveness [6.229124658686219]
  We develop a generic dual-channel detection paradigm that uses token cohesiveness as a plug-and-play module to improve existing zero-shot detectors. To calculate token cohesiveness, we use a few rounds of random token deletion and semantic difference measurement. Experiments with four state-of-the-art base detectors on various datasets, source models, and evaluation settings demonstrate the effectiveness and generality of the proposed approach.
  arXiv  Detail & Related papers  (2024-09-25T13:18:57Z)
• GlitchProber: Advancing Effective Detection and Mitigation of Glitch Tokens in Large Language Models [17.633722815221983]
  Large language models (LLMs) have achieved unprecedented success in the field of natural language processing. Recent research has discovered a class of abnormal tokens in the model's vocabulary space and named them "glitch tokens" In this work, we aim to enhance the understanding of glitch tokens and propose techniques for their detection and mitigation.
  arXiv  Detail & Related papers  (2024-08-09T07:19:53Z)
• Solving Token Gradient Conflict in Mixture-of-Experts for Large Vision-Language Model [20.979790612689992]
  Mixture-of-Experts (MoE) has gained increasing attention in studying Large Vision-Language Models (LVLMs) Existing MoE methods in LVLMs encourage different experts to handle different tokens, and they usually employ a router to predict the routing of each token. This paper proposes a novel method based on token-level gradient analysis, i.e., Solving Token Gradient Conflict (STGC)
  arXiv  Detail & Related papers  (2024-06-28T13:20:17Z)
• Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling [3.873482175367558]
  In this paper, we treat the Generation of each token by Large Language Model (LLM) as a Classification (GaC) for ensembling. In experiments, we ensemble state-of-the-art LLMs on several benchmarks, including exams, mathematics and reasoning, and observe that our method breaks the existing community performance ceiling.
  arXiv  Detail & Related papers  (2024-06-18T13:17:26Z)
• Detecting Hallucinations in Large Language Model Generation: A Token Probability Approach [0.0]
  Large Language Models (LLMs) produce inaccurate outputs, also known as hallucinations. This paper introduces a supervised learning approach employing only four numerical features derived from tokens and vocabulary probabilities obtained from other evaluators. The method yields promising results, surpassing state-of-the-art outcomes in multiple tasks across three different benchmarks.
  arXiv  Detail & Related papers  (2024-05-30T03:00:47Z)
• Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference [59.91176945361035]
  We introduce Visual Tokens Withdrawal (VTW), a plug-and-play module to boost MLLMs for rapid inference. Our approach is inspired by two intriguing phenomena we have observed. Our VTW approach can cut computational overhead by over 40% across diverse multimodal tasks while maintaining performance.
  arXiv  Detail & Related papers  (2024-05-09T14:38:53Z)
• 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)
• Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus [99.33091772494751]
  Large Language Models (LLMs) have gained significant popularity for their impressive performance across diverse fields. LLMs are prone to hallucinate untruthful or nonsensical outputs that fail to meet user expectations. We propose a novel reference-free, uncertainty-based method for detecting hallucinations in LLMs.
  arXiv  Detail & Related papers  (2023-11-22T08:39:17Z)
• Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information [67.78183175605761]
  Large Language Models are susceptible to adversarial prompt attacks. This vulnerability underscores a significant concern regarding the robustness and reliability of LLMs. We introduce a novel approach to detecting adversarial prompts at a token level.
  arXiv  Detail & Related papers  (2023-11-20T03:17:21Z)
• Hot or Cold? Adaptive Temperature Sampling for Code Generation with Large Language Models [54.72004797421481]
  We conduct the first systematic study to explore a decoding strategy specialized in code generation. Inspired by the above findings, we propose a simple yet effective method: Adaptive Temperature (AdapT) sampling. Results show that AdapT sampling significantly outperforms state-of-the-art decoding strategy.
  arXiv  Detail & Related papers  (2023-09-06T06:27:33Z)

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.