The First Workshop on Learning with Natural Language Supervision

Schedule (all times GMT+1, see Underline for zoom link)

  • 9--9:30: Opening Remarks

  • 9:30--10: Invited talk: Hinrich Schütze

  • 10--10:30: Coffee break

  • 10:30--11: Spotlight talks:

    • Rakesh Menon: CLUES: A Benchmark for Learning Classifiers using Natural Language Explanations

    • Shikhar Murthy: Fixing Model Bugs with Natural Language Patches

    • Peter Hase: When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data

  • 11--12: Breakout session 1: Datasets

  • 12--12:30: Invited talk: Jeannette Bohg

  • 12:30--14:00: Lunch

  • 14:00--15:00: Posters

  • 15:00--15:30: Coffee Break

  • 15:30--17:00: Invited talks:

    • Justin Johnson

    • Anna Ivanova

    • Hanna Hajishirzi

  • 17:00--18:00: Breakout session 2: Datasets

ACL 2022

To a growing extent, advances across machine learning application domains are driven by advances in NLP. In computer vision, image captions are used to shape learned representations of images [Frome et al., 2013, Mu et al., 2020, Radford et al., 2021, Desai and Johnson 2021]. In programming languages, textual code comments are used to guide and constrain models for example-based program synthesis [Yaghmazadeh et al., 2017, Austin et al., 2021, Wong et al., 2021]. In robotics and more general policy learning settings, rules and instructions are used to enable generalization to new environments and goals [Zhong et al., 2020, Narasimhan et al., 2018, Sharma et al., 2020]. Within NLP, rich natural-language annotations and task descriptions are used to improve the performance and interpretability of models for text categorization and question answering [Hancock et al., 2018, Weller et al., 2020, Efrat et al., 2020]. And in cognitive science, experimental evidence suggests that language shapes many other aspects of human cognition (e.g. Jones et al., 1991).

At present, however, most research on learning from language takes place within individual application domains (and mostly outside of the NLP community). While many approaches to language supervision are domain-general, and closely connected to “core” NLP research, there are currently no venues where researchers from across the field can meet to share ideas and draw connections between their disparate lines of research. Our workshop will offer a central meeting point for research on language-based supervision, enabling researchers within and beyond NLP to discuss how language processing models and algorithms can be brought to bear on problems beyond the textual realm (e.g. visual recognition, robotics, program synthesis, sequential decision making). Existing workshops like RoboNLP, SPLU, and ViGiL focus on models for multi-modality; inspired by the relationship between language and human cognitive development, our workshop will emphasize broader use of language not just as an input modality but a fundamental source of information about the structure of tasks and problem domains.

In keeping with this interdisciplinary focus, our proposed format differs in two ways from a standard NLP workshop: first, with a special emphasis on speakers and attendees who would not typically attend NLP conferences; second, by replacing the standard panel discussion with a series of workshop-wide breakout sessions aimed at seeding cross-institutional collaborations around new tasks, datasets, and models.

We accept both archival and non-archival submissions, via ACL Rolling Review or direct submission to the workshop.

Submission page:
(See details below regarding ARR.)

List of topics

We plan to offer separate archival and non-archival tracks. Relevant topics for submissions include (but are not limited to):

  • Language-based task specifications (e.g. for classification or interactive decision-making).

  • Language-guided search heuristics (e.g. for planning or theorem proving).

  • Learning from user feedback in conversational agents.

  • Generating and learning from natural language explanations and rationales.

  • Understanding and improving prompt design (in NLP and beyond).

  • Language guidance for improving safety and robustness.

  • Language-guided meta-learning.

  • Computational models of interactions between language and other aspects of cognition (e.g., categorization and memory).


  • Hinrich Schütze (LMU Munich, NLP)

  • Jeannette Bohg (Stanford, robotics)

  • Işıl Dillig (UT Austin, programs)

  • Justin Johnson (Univ. of Michigan, vision)

  • Anna Ivanova (MIT, neuroscience)

  • Hannaneh Hajishirzi (Univ. of Washington, NLP & vision)


  • Jacob Andreas (MIT)

  • Karthik Narasimhan (Princeton)

  • Aida Nematzadeh (Deepmind)

Dates and Deadlines

  • For papers submitted directly to the workshop:

    • Submission site opens: January 15, 2022

    • Submission deadline: Feb 28, 2022 March 7, 2022 11:59pm anywhere on earth (AOE)

    • Reviews due: March 21, 2022 March 28, 2022

  • For papers submitted via ACL rolling review:

    • Submission deadline (to ARR): February 15, 2022

    • Reviews due (from ARR): March 15

    • Commitment deadline: March 21

  • Decisions announced: March 26, 2022 April 3, 2022

  • Camera-ready papers due: April 10, 2022

  • Day of workshop: May 26, 2022

Submission page:
(Use the same link to commit a paper that already has ARR reviews attached.)

Submission policy

  • Submissions should be a maximum of four pages, plus any number of pages for references or appendices.

  • Submissions should be fully anonymized for double-blind review.

  • Submissions should use the ACL 2022 style file.

  • Dual submission: for our non-archival track only, we accept submissions that are under review in other venues.