BEGIN:VCALENDAR
CALSCALE:GREGORIAN
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Drupal iCal API//EN
X-WR-TIMEZONE:America/New_York
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
DTSTART:20070311T020000
TZNAME:EDT
TZOFFSETTO:-0400
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
DTSTART:20071104T020000
TZNAME:EST
TZOFFSETTO:-0500
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SEQUENCE:1
X-APPLE-TRAVEL-ADVISORY-BEHAVIOR:AUTOMATIC
224081
20250916T125900Z
DTSTART;TZID=America/New_York:20250929T120000
DTEND;TZID=America/New_York:2
 0250929T125000
URL;TYPE=URI:/news/calendar/events/bme-s
 eminar-series-bashima-islam-phd-wpi-ece-mindfulness-sensing-multimodal-rea
 soning-toward-sensor
BME Seminar Series: Bashima Islam, PhD: WPI ECE: 鈥淔rom Mindfulness Sensin
 g to Multimodal Reasoning: Toward Sensor-Language Intelligence Beyond Visi
 on鈥?
\n\n\n      \n      \n\n\n\nSeminar Series \n鈥淔rom Mindfulness Sensing to Multimodal Reasoning:
  Toward Sensor-Language Intelligence Beyond Vision鈥漒n\n\n\n\n      \n   
    \n\n\n\nBashima Islam, PhD\nAssistant Professor of Electrical and Compu
 ter Engineering\nWorcester Polytechnic Institute\n\nAbstract: The rapid gr
 owth of multimodal AI opens new opportunities for sensing, reasoning, and 
 interaction, yet most systems still focus narrowly on vision and overlook 
 signals critical for human-centered applications. In this talk, I will pre
 sent three recent projects from my group that collectively broaden the sco
 pe of multimodal intelligence. First, I will introduce our work on mindful
 ness and respiration sensing, where we design smartphone-based algorithms 
 that track respiration rate and estimate mindfulness skill progression usi
 ng only accelerometer data. This study demonstrates how sensor feedback ca
 n enhance usability and engagement in digital mindfulness training, provid
 ing a compelling case for health-oriented multimodal AI. Next, I will pres
 ent RAVEN, a unified architecture for multimodal question answering. At it
 s core is QuART, a query-conditioned token gating module that learns to as
 sign relevance scores across modalities, enabling the system to amplify in
 formative cues while suppressing distractors. Through a staged training pi
 peline, RAVEN achieves robust reasoning across video, audio, and sensor st
 reams. Finally, I will discuss LLaSA, the Large Language and Sensor Assist
 ant, which extends multimodal research to wearable sensing. LLaSA introduc
 es new datasets and evaluation frameworks for aligning sensor signals with
  language, offering the first general-purpose assistant that reasons joint
 ly over sensor data and natural language queries. Together, these projects
  chart a path toward sustainable multimodal systems that hear, sense, and 
 reason with the world, advancing both human-centered applications and core
  AI methods.\nBio: Dr. Bashima Islam is an Assistant Professor of Electric
 al and Computer Engineering at Worcester Polytechnic Institute, with affil
 iations in Computer Science and Data Science. Her research transforms AI t
 hrough the fusion of sensors, speech, and language to build sustainable, i
 ntelligent systems for the edge. She focuses on bridging AI with real-worl
 d impact that advance acoustic understanding, behavioral health monitoring
 , and real-world multimodal intelligence while considering resource constr
 ained of low-power devices. She has been recognized with several prestigio
 us honors, including the NSF CRII Award, multiple NIH research grants, and
  selection to Forbes 30 Under 30 in Science.\nFor a zoom link please conta
 ct Kate Harrison at kharrison@wpi.edu\n
END:VEVENT
END:VCALENDAR