mlsanityData to 14 Oct 2026

Both Direct and Indirect Evidence Contribute to Dative Alternation Preferences in Language Models

arXiv 2503.20850 ↗First posted 26 Mar 2025cs.CLcited by 18

Not on the map. Why?

Abstract (arXiv)

Language models (LMs) tend to show human-like preferences on a number of syntactic phenomena, but the extent to which these are attributable to direct exposure to the phenomena or more general properties of language is unclear. We explore this with the English dative alternation (DO: "gave Y the X" vs. PO: "gave the X to Y"), using a controlled rearing paradigm wherein we iteratively train small LMs on systematically manipulated input. We focus on two properties that affect the choice of alternant: length and animacy. Both properties are directly present in datives but also reflect more global tendencies for shorter elements to precede longer ones and animates to precede inanimates. First, by manipulating and ablating datives for these biases in the input, we show that direct evidence of length and animacy matters, but easy-first preferences persist even without such evidence. Then, using LMs trained on systematically perturbed datasets to manipulate global length effects (re-linearizing sentences globally while preserving dependency structure), we find that dative preferences can emerge from indirect evidence. We conclude that LMs' emergent syntactic preferences come from a mix of direct and indirect sources.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Builds on 14 of its 35 references that are in the map. First posted on arXiv 26 Mar 2025.

Builds on

14 of 35 references.

References · 14

  1. 2024Language Models Learn Rare Phenomena from Less Rare Phenomena: The Case of the Missing AANNscited by 65
  2. 2022minicons: Enabling Flexible Behavioral and Representational Analyses of Transformer Language Modelscited by 58
  3. 2025How Linguistics Learned to Stop Worrying and Love the Language Modelscited by 51
  4. 2024Filtered Corpus Training (FiCT) Shows that Language Models can Generalize from Indirect Evidencecited by 35
  5. 2024Mission: Impossible Language Modelscited by 32
  6. 2021Language Models Use Monotonicity to Assess NPI Licensingcited by 31
  7. 2024Is Child-Directed Speech Effective Training Data for Language Models?cited by 24
  8. 2025Can Language Models Learn Typologically Implausible Languages?cited by 23
  9. 2020Investigating representations of verb bias in neural language modelscited by 22
  10. 2023In-Contextual Gender Bias Suppression for Large Language Modelscited by 12
  11. 2024Manipulating language models' training data to study syntactic constraint learning: the case of English passivizationcited by 12
  12. 2024A systematic framework for generating novel experimental hypotheses from language modelscited by 9
2 more
  1. 2024Can Language Models Induce Grammatical Knowledge from Indirect Evidence?cited by 1
  2. 2025Language Models Largely Exhibit Human-like Constituent Ordering Preferencescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 13.

18 papers.

Sort:

By area:

  1. 2025Unpacking Let Alone: Human-Scale Models Generalize to a Rare Construction in Form but not Meaningcited by 4
  2. 2026An Empirical Study on Preference Tuning Generalization and Diversity Under Domain Shiftcited by 2
  3. 2026On the Interpretability of Whisper Encodings Using Sparse Autoencoderscited by 2
  4. 2026Function Words as Statistical Cues for Language Learningcited by 1
  5. 2026A Unified Assessment of the Poverty of the Stimulus Argument for Neural Language Modelscited by 1
  6. 2026Cross-Modal Taxonomic Generalization in (Vision-) Language Modelscited by 1
  7. 2026Do Language Models Know What Not to Say? Causal Evidence for Statistical Preemption in LLMscited by 1
  8. 2025Language models as tools for investigating the distinction between possible and impossible natural languagescited by 0
  9. 2025Large language models have learned to use languagecited by 0
  10. 2025Large language models are not about natural languagecited by 0
  11. 2025Linguists should learn to love speech-based deep learning modelscited by 0
  12. 2026Are Language Models Models?cited by 0
6 more
  1. 2026Across the Levels of Analysis: Explaining Predictive Processing in Humans Requires More Than Machine-Estimated Probabilitiescited by 0
  2. 2026You Can't Fight in Here! This is BBS!cited by 0
  3. 2026Targeted Linguistic Analysis of Sign Language Models with Minimal Translation Pairscited by 0
  4. 2026Not-So-Strange Love: Language Models and Generative Linguistic Theories are More Compatible than They Appearcited by 0
  5. 2026The Holistic Storage of Verb+Up Phrases in Text-based and Audio-based Language Modelscited by 0
  6. 2026Disentangling Statistical Preemption from Entrenchment in Language Models' Avoidance of Overgeneralizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2018Do RNNs learn human-like abstract word order preferences?
  2. 2026Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models
  3. 2026Preferences for Idiomatic Language are Acquired Slowly -- and Forgotten Quickly: A Case Study on Swedish
  4. 2026Language Models Generalize to Human-like Word Order Preferences
  5. 2024What Goes Into a LM Acceptability Judgment? Rethinking the Impact of Frequency and Length
  6. 2024Black Big Boxes: Tracing Adjective Order Preferences in Large Language Models
  7. 2026Left-Branching Transformers Excel at Right-Branching Languages: Data Shapes Word Order Preferences in Language Models
  8. 2026A Dataset for Probing Translationese Preferences in English-to-Swedish Translation