Papers by Matthias Scheutz
Sensitivity to Input Order: Evaluation of an Incremental and Memory-Limited Bayesian Cross-Situational Word Learning Model (C18-1)
Copied to clipboard
| Challenge: | Existing word learning models are insensitive to input order effects, but they are noisy and only provide ambiguous information. |
| Approach: | They propose a Bayesian cross-situational word learning model with an incremental memory-limited algorithm for predicting input order effects. |
| Outcome: | The proposed model performs well on corpus data while being insensitive to input order effects. |
Automating Dataset Production Using Generative Text and Image Models (2024.lrec-main)
Copied to clipboard
| Challenge: | a lack of benchmarks or data for natural language processing is hindering empirical methods . a new pipeline is proposed to reduce the burden of producing image and text datasets . |
| Approach: | They propose a pipeline to reduce the research burden of producing image and text datasets when datasets may not exist. |
| Outcome: | The proposed pipeline reduces the research burden of producing image and text datasets when datasets may not exist. |
Developing a Corpus of Indirect Speech Act Schemas (2020.lrec-1)
Copied to clipboard
| Challenge: | Indirect speech acts (ISAs) involve utterances whose literal meanings are not identical to their intended meanings. |
| Approach: | They propose a formal representation of ISA Schemas required for such testing, including a measure of the difficulty of a particular schema. |
| Outcome: | The proposed model minimizes the amount of expert authoring needed and maximizes realism. |
Reasoning Requirements for Indirect Speech Act Interpretation (2020.coling-main)
Copied to clipboard
| Challenge: | Existing systems that pretrain word and sentence embeddings to account for nearby linguistic context are unclear how to integrate extra-linguistic context into NLU. |
| Approach: | They perform a corpus analysis to develop a representation of the knowledge and reasoning used to interpret indirect speech acts. |
| Outcome: | The proposed model is based on the domain-general patterns of reasoning involved and implements Answer Set programming. |
Social Norms Guide Reference Resolution (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing tools for natural language resolution fail to handle ambiguous referents . ambiguity arises when the language is underspecified or there are multiple candidate referent. |
| Approach: | They investigate how pragmatic modulators outside of the linguistic content are critical for correct interpretation of referents in underspecified contexts. |
| Outcome: | The proposed method can be used to resolve referents in human environments. |
Towards a Conversation-Analytic Taxonomy of Speech Overlap (L18-1)
Copied to clipboard
| Challenge: | a taxonomy for classifying speech overlap in natural language dialogue is presented . the scheme classifies overlap on the basis of several features, including onset point, local dialogue history, and management behavior. |
| Approach: | They propose a taxonomy for classifying speech overlap in natural language dialogue . they describe the various dimensions of the scheme and show how it was applied to a corpus of collaborative dialogue based on onset point, dialogue history, and management behavior . |
| Outcome: | The proposed taxonomy classifies overlap on the basis of onset point, dialogue history, management behavior. |