Workshop on Computational Developmental Linguistics (CDL)

The 1st Workshop on
Computational Developmental Linguistics (CDL)

July 4th, 2026
Co-located with ACL 2026, San Diego

Overview

Background

The first workshop on computational developmental linguistics (CDL) invites interdisciplinary contributions broadly in the topic of computational developmental linguistics. By computational developmental linguistics, we broadly refer to the studies of computational modeling of language acquisition and change over time, encompassing both individual learners, including humans or machines, throughout their lifetime and across populations. Particularly, we welcome submissions that study the developmental trajectories through which learners master phonology, lexicon, and syntax, as well as the idiolectal shifts that arise as speakers adapt to new communities, domains, or interlocutors. With the advent of neural language models (LMs), we now have large-scale systems that can be examined for learning dynamics, representational change, knowledge tracing, and language adaptation. Beyond static text corpora, interactive and multimodal training regimes make it possible to reverse-engineer developmental conditions, explore resource-efficient learning, and measure parallels and divergences between human and machine language acquisition. This rapidly growing area not only advances our understanding of how linguistic capabilities emerge and evolve in artificial systems, but also provides new potential for generating hypotheses about human language learning.

Scope and Goal

This workshop aims to bridge the conversation between modern machine learning and developmental linguistics. We hope to draw inspiration from both fields, identifying similarities and differences in the (im)plausibilities of comparing LMs and human language learning, to motivate better LM engineering and more rigorous computational modeling of human language acquisition. Possible comparisons include (1) pretraining dynamics vs. first language acquisition: relating the data and scaling behavior of large models to the developmental timelines, stages, and milestones documented in human language learning; (2) continual learning and post-training vs. idiolect change and second language acquisition: exploring how ongoing adaptation, fine-tuning, or domain shifts in large models mirror semantic drift and individual language change in human speakers. By bringing together researchers from machine learning, computational linguistics, cognitive science, and developmental psychology, we aim to foster cross-disciplinary dialogue and establish frameworks for building computational models and performing computational analyses of language development, enabling scientifically rigorous comparisons between humans and machines without overly anthropomorphizing LMs.

Call for Papers

Topics of Interest

The scope and topics include, but are not limited to:

Computational models for developmental linguistics. Models and formalisms for simulating first and second language acquisition, including data constraints, model architectures, training objectives, and frameworks for modeling idiolect change and semantic drift.
Learning dynamics in pretraining and post-training of LMs. Behavioral and mechanistic interpretations of how linguistic and cognitive competence emerge during pretraining, and how it evolves through post-training (e.g., finetuning, alignment, and domain adaptation). In addition to text-only LMs, we encourage approaches that integrate multiple modalities and interactive learning, as these more closely mirror human language acquisition processes.
Comparisons of developmental linguistics in humans and machines. Comparing the processes, constraints, and outcomes of language acquisition in humans and artificial learners; examining the possibilities, limitations, and methodological challenges of such comparisons; relating model learning trajectories to developmental timelines and milestones in humans; and conducting these analyses in a scientifically rigorous manner without over-anthropomorphizing LMs.
Knowledge tracing and developmental diagnostics. Developing tools and methods to track the acquisition, transformation, and retention of linguistic knowledge in LMs, including phase transition detection, benchmarks, and probes of representational change.
Applications in language education and clinical NLP. Applying insights from computational developmental linguistics to improve LM engineering, language tutoring systems, child-directed AI interaction, adaptive educational technology, and computational models of neurological communication disorders.

Submission Format

Page Limits: We welcome short papers up to 4 pages and full papers up to 8 pages, both excluding references, with 1 additional page allowed for camera-ready version to address reviewer comments.
Archival Options: This workshop is optionally archival to ACL Anthology. Please indicate your selection of an "archival submission" or a "non-archival submission" at the bottom of your OpenReview Submission Portal. The archival option will not influence the decision of your submission.

Submissions must be prepared in PDF format using the ACL-style template through OpenReview.

Submission Portal: OpenReview
Submission Portal for ARR Commitment: OpenReview (ARR Commitment)
Submission Template: Files

Important Dates

All times are AoE (Anywhere on Earth).

Submission Open: December 15th, 2025
Submission Due: March 20th, 2026
Review Due: April 22nd, 2026
Notification of Acceptance: ~~May 1st, 2026~~ May 7th, 2026
Camera-Ready Due (tentative): ~~May 15th, 2026~~ May 21st, 2026
Workshop Day: July 4th, 2026 (co-located with ACL 2026)

Confirmed Keynote Speakers (Alphabetical Order)

Brian MacWhinney

Carnegie Mellon University

Kanishka Misra

UT Austin

Mariya Toneva

Max Planck Institute

Schedule

All times are local (PDT, UTC−7). Talk format: 35 min + 5 min Q&A.

13:30--13:35

Opening Remarks

13:40--14:20

Mariya Toneva Max Planck Institute

Invited Talk

14:20--15:00

Brian MacWhinney Carnegie Mellon University

Invited Talk

15:00--15:40

Kanishka Misra UT Austin

Invited Talk

15:40--16:30

Coffee Break

16:30--18:00

Poster Session (tentative; might be updated according to ACL instructions)

Accepted Papers

Paper Authors: please see presentation guidelines here.

Archival Papers

Linguistics Theory Meets LLM: Code-Switched Text Generation via Equivalence Constrained Large Language Models
Garry Kuwanto, Chaitanya Agarwal, Genta Indra Winata, Derry Tanti Wijaya
Do Structural Priors Help Neural Language Models Learn Grammar? Evidence from Child-Scale Data
Jon-Paul Cacioli
Fine-tuned speech representations track spoken language convergence to adult models in infants and children who are deaf/hard-of-hearing
Landon Choy, Ali Sartaz Khan, Sonia Patrizi, Daisy S. Ye, Julianna Gross, Margaret Cychosz
Do Language Models Show Structural Priming Across Different Domains?
So Young Lee, Russell Scheinberg, Ameeta Agrawal
Do large language models and humans follow similar learning stages? Assessing GPT-2's order of Swedish grammar acquisition within the Processability Theory framework
Stella Lundqvist, Murathan Kurfali, Johan Sjons
On the Learnability of Syntax from Raw Speech with Autoregressive Predictive Coding
Shunsuke Kando, Yusuke Miyao
Modeling Writing Development as Coordinated Change Across Linguistic and Semantic Dimensions
Michelle Banawan, Andrew Potter, Tracy Arner, Danielle S. McNamara
L1 Influence in L2 Language Models: A Human-centric Approach
Laura Barbenel, Lily Goulder, Aoife O'Driscoll, Suchir Salhan, Catherine Arnett, Andrew Caines, Paula Buttery
A Scalable Tool for Measuring Manner and Result Verbs in Developmental Language Research
Divyesh Pratap Singh, Dakshesh Gusain, Federica Bulgarelli, Alison Eisel Hendricks, John Beavers, Nathan M. Beers, Ifeoma Nwogu
Making Synthetic Questions More Child-Directed: Prompting and Sampling Effects
Whitney Poh, Michael Tombolini, Libby Barak

Non-Archival Papers

Discovering Properties of Inflectional Morphology in Neural Emergent Communication
Miles Gilberti, Shane Storks, Huteng Dai
Reduplication as an Inductive Bias in Pre-pretrained Language Models
Shunjie Wang
Tracking the emergence of linguistic structure in self-supervised models learning from speech
Marianne de Heer Kloots, Martijn Bentum, Hosein Mohebbi, Charlotte Pouw, Gaofei Shen, Willem Zuidema
Developmental Trajectories of Situation Modeling and Mentalizing in Transformer Language Models
Pamela D. Rivière, Cameron Robert Jones, Sean Trott
Comparing forced aligners on young child speech and examining what acoustic features control performance
Caroline Hall-Sherr, Kaitlyn Chou, Sarah C. Creel
Whisper surpasses wav2vec2 in transcribing young child speech, but neither performs well
Sarah C. Creel
CAIT: A Syntactic Parsing Toolkit for Child–Adult InTeractions
Francesca Padovani, Xiulin Yang, Bastian Bunzeck, Jaap Jumelet, Yevgen Matusevych, Nathan Schneider, Arianna Bisazza
Spectral Archaeology: The Causal Topology of Model Evolution
Valentin Noël
Child-directed speech facilitates production, not comprehension, in BabyLMs
Bastian Bunzeck, Sina Zarrieß
The Constructional Complexity Analyzer: Creating and Validating a Python Package for Measuring Syntactic Complexity Indices in Language Development
Haerim Hwang
Word-Centered Semantic Graphs for Interpretable Diachronic Sense Tracking
Imene Kolli, Kai-Robin Lange, Jonas Rieger, Carsten Jentsch

Workshop Venue

Grand Hyatt Manchester San Diego

Room: Harbor G (2nd Floor)

1 Market Pl, San Diego, CA 92101

Frequently Asked Questions

Can I submit a manuscript that is not accepted yet but is on arxiv and/or is currently under review?

Yes, please select the non-archival option. Note that computer vision conferences such as CVPR and ICCV consider peer-reviewed workshop papers as publications if their length exceeds 4 pages (excluding references), even if they do not appear in proceedings.
If you are considering submitting your paper later to these conferences, please consider submitting an abridged version as a short paper (up to 4 pages).