Constructions Are So Difficult That Even Large Language Models Get Them Right for the Wrong Reasons

• URL: http://arxiv.org/abs/2403.17760v2
• Date: Wed, 29 May 2024 23:41:37 GMT
• Title: Constructions Are So Difficult That Even Large Language Models Get Them Right for the Wrong Reasons
• Abstract summary: We introduce a small challenge dataset for NLI with large lexical overlap. We show that GPT-4 and Llama 2 fail it with strong bias. From a Computational Linguistics perspective, we identify a group of constructions with three classes of adjectives which cannot be distinguished by surface features.
• Score: 43.708431369382176
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In this paper, we make a contribution that can be understood from two perspectives: from an NLP perspective, we introduce a small challenge dataset for NLI with large lexical overlap, which minimises the possibility of models discerning entailment solely based on token distinctions, and show that GPT-4 and Llama 2 fail it with strong bias. We then create further challenging sub-tasks in an effort to explain this failure. From a Computational Linguistics perspective, we identify a group of constructions with three classes of adjectives which cannot be distinguished by surface features. This enables us to probe for LLM's understanding of these constructions in various ways, and we find that they fail in a variety of ways to distinguish between them, suggesting that they don't adequately represent their meaning or capture the lexical properties of phrasal heads.

Related papers

• Evaluating Chinese Ambiguity Understanding in Large Language Models [15.610863020079327]
  CHA-Gen is the first PA Theory-grounded Chinese ambiguity dataset.<n>It comprises 5,712 sentences (2,414 ambiguous, 3,298 unambiguous) across 18 potential ambiguous structures.
  arXiv  Detail & Related papers  (2026-05-15T05:35:18Z)
• Correct-Detect: Balancing Performance and Ambiguity Through the Lens of Coreference Resolution in LLMs [8.879888552904598]
  Large Language Models (LLMs) are intended to reflect human linguistic competencies.<n>We show that LLMs can achieve good performance with minimal prompting in both coreference disambiguation and the detection of ambiguity in coreference.<n>We present the CORRECT-DETECT trade-off: though models have both capabilities and deploy them implicitly, successful performance balancing these two abilities remains elusive.
  arXiv  Detail & Related papers  (2025-09-17T22:12:30Z)
• Drivel-ology: Challenging LLMs with Interpreting Nonsense with Depth [21.092167028989632]
  Drivelology is a linguistic phenomenon characterised by "nonsense with depth"<n>We construct a benchmark dataset of over 1,200+ meticulously curated and diverse examples across English, Mandarin, Spanish, French, Japanese, and Korean.<n>We find that current large language models (LLMs) consistently fail to grasp the layered semantics of Drivelological text.
  arXiv  Detail & Related papers  (2025-09-04T03:58:55Z)
• Transforming Hidden States into Binary Semantic Features [0.0]
  We propose to employ the distributional theory of meaning once again. Using Independent Component Analysis to overcome some of its challenging aspects, we show that large language models represent semantic features in their hidden states.
  arXiv  Detail & Related papers  (2024-09-29T22:23:52Z)
• Seemingly Plausible Distractors in Multi-Hop Reasoning: Are Large Language Models Attentive Readers? [6.525065859315515]
  We investigate whether Large Language Models (LLMs) are prone to exploiting simplifying cues in multi-hop reasoning benchmarks. Motivated by this finding, we propose a challenging multi-hop reasoning benchmark, by generating seemingly plausible multi-hop reasoning chains. We find that their performance to perform multi-hop reasoning is affected, as indicated by up to 45% relative decrease in F1 score when presented with such seemingly plausible alternatives.
  arXiv  Detail & Related papers  (2024-09-08T19:22:58Z)
• Large Language Models Lack Understanding of Character Composition of Words [3.9901365062418317]
  Large language models (LLMs) have demonstrated remarkable performances on a wide range of natural language tasks. We show that most of them fail to reliably carry out even the simple tasks that can be handled by humans with perfection.
  arXiv  Detail & Related papers  (2024-05-18T18:08:58Z)
• FAC$^2$E: Better Understanding Large Language Model Capabilities by Dissociating Language and Cognition [56.76951887823882]
  Large language models (LLMs) are primarily evaluated by overall performance on various text understanding and generation tasks. We present FAC$2$E, a framework for Fine-grAined and Cognition-grounded LLMs' Capability Evaluation.
  arXiv  Detail & Related papers  (2024-02-29T21:05:37Z)
• Towards LLM-guided Causal Explainability for Black-box Text Classifiers [16.36602400590088]
  We aim to leverage the instruction-following and textual understanding capabilities of recent Large Language Models to facilitate causal explainability. We propose a three-step pipeline via which, we use an off-the-shelf LLM to identify the latent or unobserved features in the input text. We experiment with our pipeline on multiple NLP text classification datasets, and present interesting and promising findings.
  arXiv  Detail & Related papers  (2023-09-23T11:22:28Z)
• 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)
• We're Afraid Language Models Aren't Modeling Ambiguity [136.8068419824318]
  Managing ambiguity is a key part of human language understanding. We characterize ambiguity in a sentence by its effect on entailment relations with another sentence. We show that a multilabel NLI model can flag political claims in the wild that are misleading due to ambiguity.
  arXiv  Detail & Related papers  (2023-04-27T17:57:58Z)
• Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small [68.879023473838]
  We present an explanation for how GPT-2 small performs a natural language task called indirect object identification (IOI) To our knowledge, this investigation is the largest end-to-end attempt at reverse-engineering a natural behavior "in the wild" in a language model.
  arXiv  Detail & Related papers  (2022-11-01T17:08:44Z)
• Are Representations Built from the Ground Up? An Empirical Examination of Local Composition in Language Models [91.3755431537592]
  Representing compositional and non-compositional phrases is critical for language understanding. We first formulate a problem of predicting the LM-internal representations of longer phrases given those of their constituents. While we would expect the predictive accuracy to correlate with human judgments of semantic compositionality, we find this is largely not the case.
  arXiv  Detail & Related papers  (2022-10-07T14:21:30Z)

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.