Challenge: Using representational co-speech gestures, face-to-face interaction participants resolve references to objects using speech and gestures.
Approach: They propose a multimodal reference resolution task centred on representational gestures . they propose 'self-supervised' pre-training approach to gesture representation learning that grounds body movements in spoken language.
Outcome: The proposed approach aligns with expert annotations and has significant predictive power.

Similar Papers

Disambiguating Reference in Visually Grounded Dialogues through Joint Modeling of Textual and Multimodal Semantic Structures (2025.acl-long)

Copied to clipboard

Challenge: incorporating textual and multimodal reference resolution improves performance in visual-based reference resolution . Phrase grounding is a well-established task for understanding semantic relations between mentions and objects . ambiguities caused by pronouns and ellipses can arise in visually grounded dialogues .
Approach: They propose a framework that unifies textual and multimodal reference resolution by mapping mention embeddings to object embeddements and selecting mentions or objects based on their similarity.
Outcome: The proposed framework performs better in phrase grounding than other models for this task.
A Formal Analysis of Multimodal Referring Strategies Under Common Ground (2020.lrec-1)

Copied to clipboard

Challenge: a recent study has focused on multimodality in the CL/NLP community, but it has not been widely studied.
Approach: They propose to analyze mixed-modality definite referring expressions using gestures and linguistic descriptions.
Outcome: The proposed models can predict viewer judgment of referring expressions and generate more natural and informative expressions.
Towards Understanding the Relation between Gestures and Language (2022.coling-1)

Copied to clipboard

Challenge: a new study explores the relationship between gestures and language . we use contrastive learning to learn gesture embeddings .
Approach: They adapt a semi-supervised multimodal model to learn gesture embeddings using Ted talks . they show gestures are predictive of the native language of the speaker .
Outcome: The proposed model learns gesture embeddings from a multimodal dataset . it shows that gesture embeds are predictive of the native language of the speaker .
Encoding Gesture in Multimodal Dialogue: Creating a Corpus of Multimodal AMR (2024.lrec-main)

Copied to clipboard

Challenge: Abstract Meaning Representation (AMR) was designed to represent sentence meaning in English text, but recent research has explored its adaptation to broader domains, including documents, dialogues, spatial information, cross-lingual tasks, and gesture.
Approach: They propose to annotate a multimodal (speech and gesture) AMR corpus in a task-based setting and capture coreference relationships across modalities.
Outcome: The proposed corpus captures coreference relationships across modalities, enabling fine-grained analysis of how gesture and natural language interact.
Modeling Turn-Taking with Semantically Informed Gestures (2026.findings-eacl)

Copied to clipboard

Challenge: Existing computational models of turn-taking relied on verbal cues and prosody.
Approach: They propose a framework that integrates text, audio, and gestures to model multimodal turn-taking using semantic annotations.
Outcome: The proposed framework shows that incorporating semantically guided gestures yields consistent performance gains over baselines.
J-CRe3: A Japanese Conversation Dataset for Real-world Reference Resolution (2024.lrec-main)

Copied to clipboard

Challenge: Existing studies have ground referential expressions in language to real-world objects for cooperative action generation.
Approach: They propose a Japanese Conversation dataset for real-world reference resolution that ground referential expressions to visual information observed in egocentric views.
Outcome: The proposed dataset contains egocentric video and dialogue audio of real-world conversations between two people acting as a master and assistant robot at home.
How do Multimodal Foundation Models Encode Text and Speech? An Analysis of Cross-Lingual and Cross-Modal Representations (2025.naacl-short)

Copied to clipboard

Challenge: Recent advances in foundation models have sparked growing interest in expanding their text processing capabilities to speech.
Approach: They analyze the model activations from semantically equivalent sentences across languages in the text and speech modalities and examine how text and spoken are represented in recent multimodal foundation models.
Outcome: The proposed models exhibit cross-lingual differences, but are not explicitly trained for modality-agnostic representations.
A Corpus of Natural Multimodal Spatial Scene Descriptions (L18-1)

Copied to clipboard

Challenge: Existing work on multimodal spatial descriptions combines speech and hand gestures to form a corpus of multimodal descriptions.
Approach: They present a corpus of multimodal spatial descriptions as commonly occurring in route giving tasks.
Outcome: The proposed corpus of multimodal spatial descriptions is more amenable to computational analysis and useable for learning natural computer interfaces.
Which One Are You Referring To? Multimodal Object Identification in Situated Dialogue (2023.eacl-srw)

Copied to clipboard

Challenge: a demand for multimodal dialogue systems is increasing for situated dialogues, where a dialogue agent shares a co-observed vision or physical space with the user.
Approach: They propose three methods to solve multimodal object identification problem using situated dialogue dataset SIMMC 2.1.
Outcome: The proposed method improves by 20% F1-score on the largest situated dialogue dataset, SIMMC 2.1.
Grounding Language in Multi-Perspective Referential Communication (2024.emnlp-main)

Copied to clipboard

Challenge: Using a dataset of 2,970 human-written referring expressions, we find that the performance of automated models in both reference generation and comprehension lags behind that of pairs of human agents.
Approach: They propose a task and dataset for referring expression generation and comprehension in multi-agent embodied environments where two agents must take into account one another's visual perspective to produce and understand references to objects in a scene.
Outcome: The proposed model outperforms the strongest proprietary model and improves communicative success from 58.9 to 69.3% when trained with a listener.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations