Stochastic Chameleons: Irrelevant Context Hallucinations Reveal Class-Based (Mis)Generalization in LLMs

• URL: http://arxiv.org/abs/2505.22630v2
• Date: Fri, 30 May 2025 02:10:54 GMT
• Title: Stochastic Chameleons: Irrelevant Context Hallucinations Reveal Class-Based (Mis)Generalization in LLMs
• Authors: Ziling Cheng, Meng Cao, Marc-Antoine Rondeau, Jackie Chi Kit Cheung,
• Abstract summary: We show that errors result from a structured yet flawed mechanism that we term class-based (mis)generalization.<n>Experiments on Llama-3, Mistral, and Pythia reveal that this behavior is reflected in the model's internal computations.
• Score: 36.89422086121058
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The widespread success of large language models (LLMs) on NLP benchmarks has been accompanied by concerns that LLMs function primarily as stochastic parrots that reproduce texts similar to what they saw during pre-training, often erroneously. But what is the nature of their errors, and do these errors exhibit any regularities? In this work, we examine irrelevant context hallucinations, in which models integrate misleading contextual cues into their predictions. Through behavioral analysis, we show that these errors result from a structured yet flawed mechanism that we term class-based (mis)generalization, in which models combine abstract class cues with features extracted from the query or context to derive answers. Furthermore, mechanistic interpretability experiments on Llama-3, Mistral, and Pythia across 39 factual recall relation types reveal that this behavior is reflected in the model's internal computations: (i) abstract class representations are constructed in lower layers before being refined into specific answers in higher layers, (ii) feature selection is governed by two competing circuits -- one prioritizing direct query-based reasoning, the other incorporating contextual cues -- whose relative influences determine the final output. Our findings provide a more nuanced perspective on the stochastic parrot argument: through form-based training, LLMs can exhibit generalization leveraging abstractions, albeit in unreliable ways based on contextual cues -- what we term stochastic chameleons.

Related papers

• CLATTER: Comprehensive Entailment Reasoning for Hallucination Detection [60.98964268961243]
  We propose that guiding models to perform a systematic and comprehensive reasoning process allows models to execute much finer-grained and accurate entailment decisions.<n>We define a 3-step reasoning process, consisting of (i) claim decomposition, (ii) sub-claim attribution and entailment classification, and (iii) aggregated classification, showing that such guided reasoning indeed yields improved hallucination detection.
  arXiv  Detail & Related papers  (2025-06-05T17:02:52Z)
• A Closer Look at Bias and Chain-of-Thought Faithfulness of Large (Vision) Language Models [53.18562650350898]
  Chain-of-thought (CoT) reasoning enhances performance of large language models.<n>We present the first comprehensive study of CoT faithfulness in large vision-language models.
  arXiv  Detail & Related papers  (2025-05-29T18:55:05Z)
• Computation Mechanism Behind LLM Position Generalization [59.013857707250814]
  Large language models (LLMs) exhibit flexibility in handling textual positions.<n>They can understand texts with position perturbations and generalize to longer texts.<n>This work connects the linguistic phenomenon with LLMs' computational mechanisms.
  arXiv  Detail & Related papers  (2025-03-17T15:47:37Z)
• Aggregation Artifacts in Subjective Tasks Collapse Large Language Models' Posteriors [74.04775677110179]
  In-context Learning (ICL) has become the primary method for performing natural language tasks with Large Language Models (LLMs)<n>In this work, we examine whether this is the result of the aggregation used in corresponding datasets, where trying to combine low-agreement, disparate annotations might lead to annotation artifacts that create detrimental noise in the prompt.<n>Our results indicate that aggregation is a confounding factor in the modeling of subjective tasks, and advocate focusing on modeling individuals instead.
  arXiv  Detail & Related papers  (2024-10-17T17:16:00Z)
• A hierarchical Bayesian model for syntactic priming [5.765747251519448]
  The effect of syntactic priming exhibits three well-documented empirical properties. We show how these three phenomena can be reconciled in a general learning framework. We also discuss the model's implications for the lexical basis of syntactic priming.
  arXiv  Detail & Related papers  (2024-05-24T22:26:53Z)
• Evaluating Consistency and Reasoning Capabilities of Large Language Models [0.0]
  Large Language Models (LLMs) are extensively used today across various sectors, including academia, research, business, and finance. Despite their widespread adoption, these models often produce incorrect and misleading information, exhibiting a tendency to hallucinate. This paper aims to evaluate and compare the consistency and reasoning capabilities of both public and proprietary LLMs.
  arXiv  Detail & Related papers  (2024-04-25T10:03:14Z)
• In-context Learning Generalizes, But Not Always Robustly: The Case of Syntax [36.98247762224868]
  In-context learning (ICL) is now a common method for teaching large language models (LLMs) new tasks. Do models infer the underlying structure of the task defined by the context, or do they rely on superficial generalizations that only generalize to identically distributed examples? In experiments with models from the GPT, PaLM, and Llama 2 families, we find large variance across LMs. The variance is explained more by the composition of the pre-training corpus and supervision methods than by model size.
  arXiv  Detail & Related papers  (2023-11-13T23:52:43Z)
• How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure [2.530495315660486]
  We investigate the degree to which pre-trained Transformer-based large language models represent relationships between contexts. We find that LLMs perform well in generalizing the distribution of a novel noun argument between related contexts. However, LLMs fail at generalizations between related contexts that have not been observed during pre-training.
  arXiv  Detail & Related papers  (2023-11-08T18:58:43Z)
• Simple Linguistic Inferences of Large Language Models (LLMs): Blind Spots and Blinds [59.71218039095155]
  We evaluate language understanding capacities on simple inference tasks that most humans find trivial. We target (i) grammatically-specified entailments, (ii) premises with evidential adverbs of uncertainty, and (iii) monotonicity entailments. The models exhibit moderate to low performance on these evaluation sets.
  arXiv  Detail & Related papers  (2023-05-24T06:41:09Z)

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.