Can Language Models Be Tricked by Language Illusions? Easier with Syntax, Harder with Semantics

• URL: http://arxiv.org/abs/2311.01386v2
• Date: Sun, 4 Feb 2024 15:00:11 GMT
• Title: Can Language Models Be Tricked by Language Illusions? Easier with Syntax, Harder with Semantics
• Abstract summary: We investigate LMs' more subtle judgments associated with "language illusions" We look at three illusions: the comparative illusion, the depth-charge illusion, and the negative polarity item (NPI) illusion. We show that LMs are limited both in their construal as cognitive models of human language processing and in their capacity to recognize nuanced but critical information in complicated language materials.
• Score: 7.759719313292493
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Language models (LMs) have been argued to overlap substantially with human beings in grammaticality judgment tasks. But when humans systematically make errors in language processing, should we expect LMs to behave like cognitive models of language and mimic human behavior? We answer this question by investigating LMs' more subtle judgments associated with "language illusions" -- sentences that are vague in meaning, implausible, or ungrammatical but receive unexpectedly high acceptability judgments by humans. We looked at three illusions: the comparative illusion (e.g. "More people have been to Russia than I have"), the depth-charge illusion (e.g. "No head injury is too trivial to be ignored"), and the negative polarity item (NPI) illusion (e.g. "The hunter who no villager believed to be trustworthy will ever shoot a bear"). We found that probabilities represented by LMs were more likely to align with human judgments of being "tricked" by the NPI illusion which examines a structural dependency, compared to the comparative and the depth-charge illusions which require sophisticated semantic understanding. No single LM or metric yielded results that are entirely consistent with human behavior. Ultimately, we show that LMs are limited both in their construal as cognitive models of human language processing and in their capacity to recognize nuanced but critical information in complicated language materials.

Related papers

• Tracing the ongoing emergence of human-like reasoning in Large Language Models [0.13048920509133807]
  Large Language Models show human-like performance on many tasks.<n>It remains unclear whether they reason like humans.<n>We find that humans enrich logical reasoning through pragmatic inferences across languages.
  arXiv  Detail & Related papers  (2026-05-20T15:28:52Z)
• LLMorphism: When humans come to see themselves as language models [0.0]
  I argue that the rise of conversational LLMs may make this bias increasingly psychologically available.<n>I distinguish LLMorphism from mechanomorphism, anthropomorphism, computationalism, dehumanization, objectification, and predictive-processing theories of mind.<n>I conclude that the issue is not only whether we are attributing too much mind to machines, but also whether we are beginning to attribute too little mind to humans.
  arXiv  Detail & Related papers  (2026-05-06T20:27:15Z)
• What can LLMs tell us about the mechanisms behind polarity illusions in humans? Experiments across model scales and training steps [0.0]
  I investigate if and how two well-known polarity illusions, the NPI illusion and the depth charge illusion, arise in LLMs.<n>The NPI illusion becomes weaker and ultimately disappears as model size increases, while the depth charge illusion becomes stronger in larger models.<n>It may not be necessary to assume "rational inference" mechanisms that convert ill-formed sentences into well-formed ones to explain polarity illusions.
  arXiv  Detail & Related papers  (2026-03-29T20:22:00Z)
• On the Thinking-Language Modeling Gap in Large Language Models [68.83670974539108]
  We show that there is a significant gap between the modeling of languages and thoughts.<n>We propose a new prompt technique termed Language-of-Thoughts (LoT) to demonstrate and alleviate this gap.
  arXiv  Detail & Related papers  (2025-05-19T09:31:52Z)
• Going Whole Hog: A Philosophical Defense of AI Cognition [0.0]
  We argue against prevailing methodologies in AI philosophy, rejecting starting points based on low-level computational details. We employ 'Holistic Network Assumptions' to argue for the full suite of cognitive states. We conclude by speculating on the possibility of LLMs possessing 'alien' contents beyond human conceptual schemes.
  arXiv  Detail & Related papers  (2025-04-18T11:36:25Z)
• How Deep is Love in LLMs' Hearts? Exploring Semantic Size in Human-like Cognition [75.11808682808065]
  This study investigates whether large language models (LLMs) exhibit similar tendencies in understanding semantic size. Our findings reveal that multi-modal training is crucial for LLMs to achieve more human-like understanding. Lastly, we examine whether LLMs are influenced by attention-grabbing headlines with larger semantic sizes in a real-world web shopping scenario.
  arXiv  Detail & Related papers  (2025-03-01T03:35:56Z)
• Can Language Models Learn Typologically Implausible Languages? [62.823015163987996]
  Grammatical features across human languages show intriguing correlations often attributed to learning biases in humans. We discuss how language models (LMs) allow us to better determine the role of domain-general learning biases in language universals. We test LMs on an array of highly naturalistic but counterfactual versions of the English (head-initial) and Japanese (head-final) languages.
  arXiv  Detail & Related papers  (2025-02-17T20:40:01Z)
• Non-literal Understanding of Number Words by Language Models [33.24263583093367]
  Humans naturally interpret numbers non-literally, combining context, world knowledge, and speaker intent. We investigate whether large language models (LLMs) interpret numbers similarly, focusing on hyperbole and pragmatic halo effects.
  arXiv  Detail & Related papers  (2025-02-10T07:03:00Z)
• Training Language Models on the Knowledge Graph: Insights on Hallucinations and Their Detectability [83.0884072598828]
  Hallucinations come in many forms, and there is no universally accepted definition. We focus on studying only those hallucinations where a correct answer appears verbatim in the training set. We find that for a fixed dataset, larger and longer-trained LMs hallucinate less. While we see detector size improves performance on fixed LM's outputs, we find an inverse relationship between the scale of the LM and the detectability of its hallucinations.
  arXiv  Detail & Related papers  (2024-08-14T23:34:28Z)
• Strong and weak alignment of large language models with human values [1.6590638305972631]
  Minimizing negative impacts of Artificial Intelligent (AI) systems requires them to be able to align with human values. We argue that this is required for AI systems like large language models (LLMs) to be able to recognize situations presenting a risk that human values may be flouted. We propose a new thought experiment that we call "the Chinese room with a word transition dictionary", in extension of John Searle's famous proposal.
  arXiv  Detail & Related papers  (2024-08-05T11:27:51Z)
• "I'm Not Sure, But...": Examining the Impact of Large Language Models' Uncertainty Expression on User Reliance and Trust [51.542856739181474]
  We show how different natural language expressions of uncertainty impact participants' reliance, trust, and overall task performance. We find that first-person expressions decrease participants' confidence in the system and tendency to agree with the system's answers, while increasing participants' accuracy. Our findings suggest that using natural language expressions of uncertainty may be an effective approach for reducing overreliance on LLMs, but that the precise language used matters.
  arXiv  Detail & Related papers  (2024-05-01T16:43:55Z)
• Language in Vivo vs. in Silico: Size Matters but Larger Language Models Still Do Not Comprehend Language on a Par with Humans [1.8434042562191815]
  This work investigates the role of model scaling in determining whether differences between humans and models are amenable to model size. We test three Large Language Models (LLMs) on a grammaticality judgment task featuring anaphora, center embedding, comparatives, and negative polarity. We find that humans are overall less accurate than ChatGPT-4 (76% vs. 80% accuracy, respectively), but that this is due to ChatGPT-4 outperforming humans only in one task condition, namely on grammatical sentences.
  arXiv  Detail & Related papers  (2024-04-23T10:09:46Z)
• Theory of Mind abilities of Large Language Models in Human-Robot Interaction : An Illusion? [18.770522926093786]
  Large Language Models have shown exceptional generative abilities in various natural language and generation tasks. We study a special application of ToM abilities that has higher stakes and possibly irreversible consequences. We focus on the task of Perceived Behavior Recognition, where a robot employs a Large Language Model (LLM) to assess the robot's generated behavior in a manner similar to human observer.
  arXiv  Detail & Related papers  (2024-01-10T18:09:36Z)
• Divergences between Language Models and Human Brains [63.405788999891335]
  Recent research has hinted that brain signals can be effectively predicted using internal representations of language models (LMs) We show that there are clear differences in how LMs and humans represent and use language. We identify two domains that are not captured well by LMs: social/emotional intelligence and physical commonsense.
  arXiv  Detail & Related papers  (2023-11-15T19:02:40Z)
• Roles of Scaling and Instruction Tuning in Language Perception: Model vs. Human Attention [58.817405319722596]
  This work compares the self-attention of several large language models (LLMs) in different sizes to assess the effect of scaling and instruction tuning on language perception. Results show that scaling enhances the human resemblance and improves the effective attention by reducing the trivial pattern reliance, while instruction tuning does not. We also find that current LLMs are consistently closer to non-native than native speakers in attention, suggesting a sub-optimal language perception of all models.
  arXiv  Detail & Related papers  (2023-10-29T17:16:40Z)
• Analyzing and Mitigating Object Hallucination in Large Vision-Language Models [110.12460299261531]
  Large vision-language models (LVLMs) have shown remarkable abilities in understanding visual information with human languages. LVLMs still suffer from object hallucination, which is the problem of generating descriptions that include objects that do not actually exist in the images. We propose a powerful algorithm, LVLM Hallucination Revisor (LURE), to rectify object hallucination in LVLMs by reconstructing less hallucinatory descriptions.
  arXiv  Detail & Related papers  (2023-10-01T18:10:53Z)
• Human-Like Intuitive Behavior and Reasoning Biases Emerged in Language Models -- and Disappeared in GPT-4 [0.0]
  We show that large language models (LLMs) exhibit behavior that resembles human-like intuition. We also probe how sturdy the inclination for intuitive-like decision-making is.
  arXiv  Detail & Related papers  (2023-06-13T08:43:13Z)
• Large Language Models are In-Context Semantic Reasoners rather than Symbolic Reasoners [75.85554779782048]
  Large Language Models (LLMs) have excited the natural language and machine learning community over recent years. Despite of numerous successful applications, the underlying mechanism of such in-context capabilities still remains unclear. In this work, we hypothesize that the learned textitsemantics of language tokens do the most heavy lifting during the reasoning process.
  arXiv  Detail & Related papers  (2023-05-24T07:33:34Z)
• Thinking Fast and Slow in Large Language Models [0.08057006406834465]
  Large language models (LLMs) are currently at the forefront of intertwining AI systems with human communication and everyday life. In this study, we show that LLMs like GPT-3 exhibit behavior that resembles human-like intuition - and the cognitive errors that come with it.
  arXiv  Detail & Related papers  (2022-12-10T05:07:30Z)

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.