Linguistic Principles for
Foundation Models

Rephrase a problem, and a model's reasoning can collapse. What if language itself is the design axis we've been missing?

NeurIPS 2026 | Dec 12, 2026 | Sydney, Australia

Workshop Overview

We extend our submission deadline to 6 Sep 2026 AoE, submit your research right now!

📢 We are actively recruiting reviewers and sponsors — if you are interested, please reach out at yangzqccc@gmail.com.

The scaling paradigm is approaching its limits on reasoning-intensive tasks[1], and agentic and scientific applications increasingly push models into domains where free-form natural language is demonstrably inadequate[2]. This workshop treats the linguistic[3][4] and symbolic[5] medium through which foundation models perceive, reason, and act as a first-class design axis, on par with architecture and scale.

Call for Papers

We invite high-quality submissions at the intersection of linguistic principles and foundation models. Accepted papers will be presented as posters, and selected submissions will be invited for contributed spotlight talks.

Submission Guidelines

Submissions must be in English and use the NeurIPS 2026 workshop LaTeX template. Papers are submitted as a single PDF:

  • Full Papersat most 9 pages (main text)
  • Short Papersat most 4 pages (main text)
  • Demo Tracklive demonstrations of systems and tools that put linguistic principles to work, presented alongside the poster sessions (4 pages max in main text)
  • Position Paperson the role of linguistic and symbolic media in the design, evaluation, and societal impact of foundation models (9 pages max in main text)

References and appendices do not count toward the page limit, but the main text must be self-contained.

  • ReviewDouble-blind. Submissions must be fully anonymized.
  • PortalManaged through OpenReview
  • Non-archivalAuthors retain full copyright; extended versions may be submitted elsewhere. Under-review NeurIPS papers are welcome.

Important Dates

  1. Jul 18, 2026
    Call for Papers opens
  2. Sep 6, 2026
    Submission deadline
    Non-archival; under-review NeurIPS papers are welcome and will receive full consideration.
  3. Sep 6–26, 2026
    Review period
  4. Sep 29, 2026
    Author notification
    AoE · NeurIPS mandatory deadline
  5. Oct 10, 2026
    Camera-ready
    AoE
  6. Dec 12, 2026
    Workshop day
    NeurIPS 2026 · Sydney, Australia

Topics of Interest

Topics include, but are not limited to, the following.

i

Language Representation Design

Structured, symbolic, or formal representations — including typed slots, intermediate notations, and formal grammars — that shape model schemas and improve reasoning, planning, and tool use.

ii

Compositionality, Semantics, and Pragmatics

How foundation models compose meaning from parts and generalize to novel combinations, and how reference, presupposition, implicature, scope, modality, and discourse structure are handled.

iii

Typology, Multilinguality, Morphology, and Tokenization

Cross-linguistic variation, low-resource languages, language-universal versus language-specific representations, and how tokenization and morphological complexity shape capability, including principled alternatives to current schemes.

iv

Psycholinguistics and Language Acquisition

Foundation models as learners and cognitive models, with comparisons to human acquisition and processing. Bridging developmental psycholinguistics and machine learning.

v

Probing, Interpretability, and Emergent Linguistic Structure

Mechanistic analysis of how syntactic, semantic, and pragmatic structures are encoded in model internals: how linguistic structure emerges, is represented, and can be recovered from neural computations.

vi

Reasoning, Agents, and Formal / Scientific Languages

How paraphrases, formal rewrites, and intermediate languages (code, math, logic) reshape reasoning on the same task; structured languages for agent planning and tool use; and domain-specific notations such as Lean, PDDL, and LTL as media of model reasoning.

vii

Linguistically Grounded Evaluation

Benchmarks beyond surface accuracy, covering compositional generalization, paraphrase robustness, pragmatic inference, and cross-linguistic competence.

viii

Beyond Text, Self-Designed Languages, and Theory

Extending linguistic principles to vision-language, speech, sign-language, and embodied foundation models; models that design their own symbolic media; and theoretical frameworks linking language structure, schema formation, and model capability.

Schedule

Saturday, 12 December 2026 · International Convention Centre Sydney · MR C3.3.
All times are Sydney local time (AEDT, UTC+11). Oral presentations are 6 minutes each, including Q&A.

Time Speaker Topic
09:00–09:10 Organizers Opening Remarks
09:10–09:40 Wray Buntine Linguistic and Interpretative Capabilities of Foundation Models for Language
09:40–10:10 Cécile Paris Question Answering Fit for Purpose: A Perspective from Natural Language Processing and User Modeling
10:10–10:40 Eduard Hovy The Structure of Semantic Representation in Embedding Space
Contributed Orals I · 6 minutes each, including Q&A
10:40–10:46 Emanuele Sansone Can AI Understand the Language of Origami?
10:46–10:52 Nicholas Runcie Talk Like a Molecular Graph: Chemical Representations for Large Language Models
10:52–10:58 Jindong Wang Beyond Structural Unification: A Perspective and Roadmap on Unified Multimodal Models
10:58–11:04 Kaibang Zhang Same Answer, Different Path: Measuring Prompt Sensitivity Beyond Task Outcomes in Small Language Models
11:04–11:10 Tim Tomov What Is a Clarification? Estimating Model-Dependent Input Ambiguity
11:10–11:16 Han Zhang Adaptive Content-Contrastive Decoding for Representation-Robust Tabular Reasoning
11:16–11:36 — Coffee Break
Contributed Orals II · 6 minutes each, including Q&A
11:36–11:42 Yan Liu Triplet-Aware Sparse Higher-Order Attention for Natural Language Understanding
11:42–11:48 Kai Han Where Natural Language Ends in Agents: Executable Skill Composition with MetaSkill
11:48–11:54 Chaoyang Zhou Intra-Trajectory Consistency for Reward Modeling
11:54–12:00 Daniel Yoo The Assistant Persona Differs Across Languages in Activation Space
12:00–12:06 Benedict Tessler Model Hypnosis: Strong control of AI via additive subliminal effects
12:06–13:00 — Lunch Break
13:00–13:40 Roger Levy Online Linguistic structure, probability, and language processing in humans and large language models
13:40–14:10 Jey Han Lau How linguistic relationships shape representation
14:10–14:40 Tal Linzen Less Superhuman Language Models
14:40–15:10 Usman Naseem Mechanistic-Driven AI Safety and Alignment
15:10–15:40 Reza Haffari What Should an Instruction Make Explicit? Language and Planning in Foundation Models
15:40–16:00 — Break
16:00–17:00 — Poster Session & Coffee Break
17:00–17:10 Organizers Closing Remarks

The programme is subject to change · a shareable version of this page lives at lp4fm.github.io/invitations

Invited Speakers

We are excited to welcome leading researchers from linguistics, cognitive science, natural language processing, and machine learning.

Advisory Panel

Organizing Committee

Contact

Questions or feedback? Feel free to reach out.

General Inquiries

For questions about the workshop, submissions, or collaboration:

yangzqccc@gmail.com

Workshop Location

NeurIPS 2026 · International Convention Centre Sydney

Exact venue and time slot will be announced closer to the event.

Is the workshop in-person or virtual?

The workshop is held in-person at NeurIPS 2026 in Sydney. Remote presentations are supported to accommodate last-minute schedule changes for invited speakers and authors.

Are the proceedings archival?

No, the workshop proceedings are non-archival. Authors of accepted papers retain full copyright and are free to submit extended versions to future venues.

Do you offer travel or registration support?

We plan to offer a small number of registration-fee waivers and, where possible, travel subsidies, with priority for student authors and participants from under-resourced institutions.

Can I submit previously published work?

Yes. Previously published work is welcome for presentation but is not eligible for workshop awards.

References

  1. [1] Zhiqin Yang, Yuhan Liu, Jingwen Fu, Pei Fu, Bo Han, Masashi Sugiyama, and Nanning Zheng. Shaping schema via language representation as the next frontier for LLM intelligence expanding. arXiv preprint arXiv:2605.09271, 2026.
  2. [2] Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, and Samuel J. Gershman. Building machines that learn and think like people. Behavioral and Brain Sciences, 40:e253, 2017.
  3. [3] Chenxi Wang, Yixuan Zhang, Lang Gao, Zixiang Xu, Zirui Song, Yanbo Wang, and Xiuying Chen. Under the shadow of Babel: How language shapes reasoning in LLMs. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 24327–24344, 2025.
  4. [4] Zixiang Xu, Yanbo Wang, Yue Huang, Xiuying Chen, Jieyu Zhao, Meng Jiang, and Xiangliang Zhang. Cross-lingual pitfalls: Automatic probing cross-lingual weakness of multilingual large language models. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8254–8284, 2025.
  5. [5] Bailin Wang, Zi Wang, Xuezhi Wang, Yuan Cao, Rif A. Saurous, and Yoon Kim. Grammar prompting for domain-specific language generation with large language models. Advances in Neural Information Processing Systems, 36:65030–65055, 2023.