Making Sense AAAI-27
Join PC Submit

AAAI-27 Workshop

Making Sense

Foundation Models for Multisensory Intelligence

A half-day workshop on how foundation models for sensor data should be built, evaluated and trusted.

Submit via OpenReview Read the call for papers
  • February 22 or 23, 2027Final day to be confirmed
  • Montréal, CanadaPalais des congrès de Montréal
  • Half day, 14:00–18:00Room to be announced
Explore AAAI-27

01The workshop

Foundation models are
reaching the senses.

Bringing together

  • Health sensing
  • Ubiquitous computing
  • Audio
  • Robotics
  • Industrial sensing
  • Time series

Wearables stream heart rhythms, motion and sleep from hundreds of millions of wrists. Microphones and earables capture speech, breathing and coughs. Tactile and olfactory sensors let robots and assistants feel and smell. Factories, vehicles and grids emit continuous telemetry.

Sensor foundation models are now deployed to millions of users. However, they are built in silos, one per application community, and the questions that decide whether multisensory intelligence works remain open. How should heterogeneous sensors be aligned with language? How should missing data and device heterogeneity be handled at scale? Which sensors carry unique or synergistic information? Do sensor models reason or merely recognize? How can they be deployed safely on the body?

02Topics

The questions
that remain open.

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

  1. Wearable and biosignal foundation models (PPG, ECG, EEG, EMG)
  2. Sensor-language models and reasoning agents
  3. Tokenization and pretraining for heterogeneous sensors
  4. Missing modalities, irregular sampling and device heterogeneity
  5. Audio, tactile, olfactory and embodied sensing
  6. Egocentric, ambient and industrial telemetry sensing
  7. Multisensory fusion and transfer across speech, vision and sensor models
  8. Benchmarks, datasets and evaluation of sensor reasoning
  9. On-device, efficient, cost-aware and dynamic sensing
  10. Personal-health agents, clinical validation, privacy and safety on the body

03Call for papers

Papers, demos
and datasets.

Submissions use the AAAI-27 template and are reviewed double-blind, so please anonymize your paper. We accept two kinds of paper.

Track 01

Research papers

Novel methods, theory or experiments. Eligible for the Best Paper Award.

Track 02

Demo and dataset papers

New datasets, benchmarks or demonstrable systems. Eligible for the Best Demo or Dataset Award.

Long papers
Up to 7 pages of technical content, with unlimited pages for references and appendices
Short papers
Up to 4 pages of technical content, with unlimited pages for references and appendices
Supplementary material
Allowed, but reviewers are not required to review it
Awards
Judged on the quality of the work, regardless of page length

Key dates

  1. Paper submission deadline23:59 AoE
  2. Author notification
  3. Camera-ready deadlineTo be announced
  4. Workshop at AAAI-2722 or 23 Feb 2027Final day to be confirmed

Submissions are open

Submit through the Making Sense venue on OpenReview.

An OpenReview account is required. New accounts can take a few days to be activated, so please register early. Submissions close on 20 November 2026 at 23:59 AoE (21 November, 11:59 UTC).

Submit via OpenReview

04Program

One afternoon of
talks, papers and demos.

Tentative schedule

Half a day on February 22 or 23, 2027, with keynote talks, oral presentations, a poster and live demo session, and a panel discussion. The final date will be announced.

Part 1 · 14:00–16:00
–Opening remarks
–Keynote 1
–Oral session I2 papers · 10 minutes each
–Keynote 2
–Poster and demo spotlight30-second introductions
–Coffee break, posters and live demos
Part 2 · 16:00–18:00
–Keynote 3
–Oral session II2 papers · 10 minutes each
–Keynote 4
–Panel discussionDo Sensor Foundation Models Reason, and Can We Trust Them on the Body?
–Award ceremony and closing remarksBest Paper Award · Best Demo or Dataset Award

All times are local to Montréal. Session times are provisional. Speakers, talk titles and accepted papers will be announced after confirmation.

Keynotes are 30 minutes (25 minutes plus 5 minutes of questions); oral presentations are 10 minutes (8 minutes plus 2 minutes of questions).

There are no criteria to attend beyond AAAI-27 registration. Anyone interested in foundation models for sensor data and multisensory intelligence is welcome. The number of attendees depends on the size of the room.

05Speakers

Speakers & panelists
to be announced.

Profiles of the keynote speakers and panelists will be posted here once they are confirmed.

Keynote speakers

Keynote 1

To be announced

Keynote 2

To be announced

Keynote 3

To be announced

Keynote 4

To be announced

Panel discussion

Do Sensor Foundation Models Reason, and Can We Trust Them on the Body?

Panelist 1

To be announced

Panelist 2

To be announced

Panelist 3

To be announced

06Organizers

The people behind
Making Sense.

Meet the workshop chairs and the organizing committee.

07Program committee

Expertise behind
every review.

Help shape the Making Sense program. We are inviting reviewers who can give careful, constructive feedback on foundation models for sensor data and multisensory intelligence.

Join the Program Committee
Who
Researchers and practitioners in health sensing, ubiquitous computing, audio, robotics, industrial sensing and time series
Reviewing
Double-blind reviews on OpenReview. Tell us how many papers you can take when you apply.
When
Between the submission deadline on 20 November and author notification on 2 December 2026
Selection
The organizers confirm appointments based on expertise. Confirmed members will be listed here.

08Call for sponsors

Partner with
Making Sense.

We are looking for sponsors who share our interest in foundation models for sensor data and multisensory intelligence. Sponsorship helps us recognize outstanding work and bring the sensing communities together. Every partnership is shaped around your goals.

Get in touch