The deman d for more sophisticated natural human-computer an d human-robot interactions is rapidly increasing a s users become more accustomed to conversation-lik e interactions with AI and NLP systems. Such inter actions require not only the robust recognition an d generation of expressions through multiple modal ities (language, gesture, vision, action, etc.), b ut also the encoding of situated meaning.
\n\nThis workshop intends to bring together resear chers who aim to capture elements of multimodal in teraction such as language, gesture, gaze, and fac ial expression with formal semantic representation s. We provide a space for both theoretical and pra ctical discussion of how linguistic co-modalities support, inform, and align with 'meaning' found in the linguistic signal alone. MMSR 2021 is co-loca ted with IWCS 2021.
\nWe soli cit papers on multimodal semantic representation. Two types of submissions are solicited: long paper s and short papers. Long papers should describe or iginal research and must not exceed 8 pages, exclu ding references. Short papers (typically system or project descriptions, or ongoing research) must n ot exceed 4 pages, excluding references. Both type s will be published in the workshop proceedings an d in the ACL Anthology. Accepted papers get an ext ra page in the camera-ready version. We strongly e ncourage students to submit to the workshop and wi ll consider a student session depending on the num ber of submissions.
\n