Causal Interventions Reveal Shared Structure Across English Filler-Gap Constructions

• URL: http://arxiv.org/abs/2505.16002v2
• Date: Tue, 30 Sep 2025 01:57:02 GMT
• Title: Causal Interventions Reveal Shared Structure Across English Filler-Gap Constructions
• Authors: Sasha Boguraev, Christopher Potts, Kyle Mahowald,
• Abstract summary: We argue that causal interpretability methods, applied to Language Models, can greatly enhance the value of such evidence.<n>Our empirical focus is a set of English filler-gap dependency constructions.<n>We show that LMs converge on similar abstract analyses of these constructions.
• Score: 35.29123099176241
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Language Models (LMs) have emerged as powerful sources of evidence for linguists seeking to develop theories of syntax. In this paper, we argue that causal interpretability methods, applied to LMs, can greatly enhance the value of such evidence by helping us characterize the abstract mechanisms that LMs learn to use. Our empirical focus is a set of English filler-gap dependency constructions (e.g., questions, relative clauses). Linguistic theories largely agree that these constructions share many properties. Using experiments based in Distributed Interchange Interventions, we show that LMs converge on similar abstract analyses of these constructions. These analyses also reveal previously overlooked factors -- relating to frequency, filler type, and surrounding context -- that could motivate changes to standard linguistic theory. Overall, these results suggest that mechanistic, internal analyses of LMs can push linguistic theory forward.

Related papers

• Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models [77.98801218316505]
  Large language models (LLMs) exhibit emergent behaviors suggestive of human-like reasoning.<n>We investigate the internal processing of LLMs during in-context concept inference.
  arXiv  Detail & Related papers  (2026-02-08T03:14:39Z)
• On the Entity-Level Alignment in Crosslingual Consistency [62.33186691736433]
  SubSub and SubInj integrate English translations of subjects into prompts across languages, leading to substantial gains in factual recall accuracy and consistency.<n>These interventions reinforce the entity representation alignment in the conceptual space through model's internal pivot-language processing.
  arXiv  Detail & Related papers  (2025-10-11T16:26:50Z)
• Under the Shadow of Babel: How Language Shapes Reasoning in LLMs [27.48119976373105]
  We show that large language models internalize the habitual logical structures embedded in different languages.<n>Our study reveals three key findings: (1) LLMs exhibit typologically aligned attention patterns, focusing more on causes and sentence-initial connectives in Chinese, while showing a more balanced distribution in English.
  arXiv  Detail & Related papers  (2025-06-19T09:06:38Z)
• When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners [111.50503126693444]
  We show that language-specific ablation consistently boosts multilingual reasoning performance.<n>Compared to post-training, our training-free ablation achieves comparable or superior results with minimal computational overhead.
  arXiv  Detail & Related papers  (2025-05-21T08:35:05Z)
• Investigating Syntactic Biases in Multilingual Transformers with RC Attachment Ambiguities in Italian and English [1.2891210250935148]
  We investigate whether monolingual and multilingual LLMs show human-like preferences when presented with examples of relative clause attachment ambiguities in Italian and English.<n>We also test whether these preferences can be modulated by lexical factors which have been shown to be tied to subtle constraints on syntactic and semantic relations.
  arXiv  Detail & Related papers  (2025-04-14T05:19:23Z)
• Disparities in LLM Reasoning Accuracy and Explanations: A Case Study on African American English [66.97110551643722]
  We investigate dialectal disparities in Large Language Models (LLMs) reasoning tasks.<n>We find that LLMs produce less accurate responses and simpler reasoning chains and explanations for AAE inputs.<n>These findings highlight systematic differences in how LLMs process and reason about different language varieties.
  arXiv  Detail & Related papers  (2025-03-06T05:15:34Z)
• Sparse Auto-Encoder Interprets Linguistic Features in Large Language Models [40.12943080113246]
  We present a systematic and comprehensive causal investigation using sparse auto-encoders (SAEs)<n>We extract a wide range of linguistic features from six dimensions.<n>We introduce two indices-Feature Representation Confidence (FRC) and Feature Intervention Confidence (FIC)-to measure the ability of linguistic features to capture and control linguistic phenomena.
  arXiv  Detail & Related papers  (2025-02-27T18:16:47Z)
• Can Language Models Learn Typologically Implausible Languages? [62.823015163987996]
  Grammatical features across human languages show intriguing correlations often attributed to learning biases in humans.<n>We discuss how language models (LMs) allow us to better determine the role of domain-general learning biases in language universals.<n>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)
• Large Language Models as Proxies for Theories of Human Linguistic Cognition [2.624902795082451]
  We consider the possible role of current large language models (LLMs) in the study of human linguistic cognition.<n>We focus on the use of such models as proxies for theories of cognition that are relatively linguistically-neutral in their representations and learning.
  arXiv  Detail & Related papers  (2025-02-11T16:38:16Z)
• Black Big Boxes: Do Language Models Hide a Theory of Adjective Order? [5.395055685742631]
  In English and other languages, multiple adjectives in a complex noun phrase show intricate ordering patterns that have been a target of much linguistic theory. We review existing hypotheses designed to explain Adjective Order Preferences (AOPs) in humans and develop a setup to study AOPs in language models. We find that all models' predictions are much closer to human AOPs than predictions generated by factors identified in theoretical linguistics.
  arXiv  Detail & Related papers  (2024-07-02T10:29:09Z)
• Explaining Text Similarity in Transformer Models [52.571158418102584]
  Recent advances in explainable AI have made it possible to mitigate limitations by leveraging improved explanations for Transformers. We use BiLRP, an extension developed for computing second-order explanations in bilinear similarity models, to investigate which feature interactions drive similarity in NLP models. Our findings contribute to a deeper understanding of different semantic similarity tasks and models, highlighting how novel explainable AI methods enable in-depth analyses and corpus-level insights.
  arXiv  Detail & Related papers  (2024-05-10T17:11:31Z)
• How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering [52.86931192259096]
  Knowledge Base Question Answering (KBQA) aims to answer natural language questions based on facts in knowledge bases. Recent works leverage the capabilities of large language models (LLMs) for logical form generation to improve performance.
  arXiv  Detail & Related papers  (2024-01-11T09:27:50Z)
• Zero-shot Causal Graph Extrapolation from Text via LLMs [50.596179963913045]
  We evaluate the ability of large language models (LLMs) to infer causal relations from natural language. LLMs show competitive performance in a benchmark of pairwise relations without needing (explicit) training samples. We extend our approach to extrapolating causal graphs through iterated pairwise queries.
  arXiv  Detail & Related papers  (2023-12-22T13:14:38Z)
• Explanation-aware Soft Ensemble Empowers Large Language Model In-context Learning [50.00090601424348]
  Large language models (LLMs) have shown remarkable capabilities in various natural language understanding tasks. We propose EASE, an Explanation-Aware Soft Ensemble framework to empower in-context learning with LLMs.
  arXiv  Detail & Related papers  (2023-11-13T06:13:38Z)
• 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)
• "You Are An Expert Linguistic Annotator": Limits of LLMs as Analyzers of Abstract Meaning Representation [60.863629647985526]
  We examine the successes and limitations of the GPT-3, ChatGPT, and GPT-4 models in analysis of sentence meaning structure. We find that models can reliably reproduce the basic format of AMR, and can often capture core event, argument, and modifier structure. Overall, our findings indicate that these models out-of-the-box can capture aspects of semantic structure, but there remain key limitations in their ability to support fully accurate semantic analyses or parses.
  arXiv  Detail & Related papers  (2023-10-26T21:47:59Z)
• 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)
• A Theory of Emergent In-Context Learning as Implicit Structure Induction [8.17811111226145]
  Scaling large language models leads to an emergent capacity to learn in-context from example demonstrations. We argue that in-context learning relies on recombination of compositional operations found in natural language data. We show how in-context learning is supported by a representation of the input's compositional structure.
  arXiv  Detail & Related papers  (2023-03-14T15:24:05Z)

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.