From clark.3664 at buckeyemail.osu.edu Sun Oct 8 14:53:57 2023 From: clark.3664 at buckeyemail.osu.edu (Clark, Christian) Date: Sun, 8 Oct 2023 18:53:57 +0000 Subject: [CaCL] Reading for 10/19 Message-ID: Hi CaCL members, CaCL will not meet on 10/12 because of autumn break. On 10/19, we will discuss "Finding Structure in One Child's Linguistic Experience" by Wang et al. (2023). Paper: https://onlinelibrary.wiley.com/doi/abs/10.1111/cogs.13305 Zoom: https://osu.zoom.us/j/93495474520?pwd=NWFRZzF6QnZrQWdheThxbWJyNjJ4dz09 Abstract: Neural network models have recently made striking progress in natural language processing, but they are typically trained on orders of magnitude more language input than children receive. What can these neural networks, which are primarily distributional learners, learn from a naturalistic subset of a single child's experience? We examine this question using a recent longitudinal dataset collected from a single child, consisting of egocentric visual data paired with text transcripts. We train both language-only and vision-and-language neural networks and analyze the linguistic knowledge they acquire. In parallel with findings from Jeffrey Elman's seminal work, the neural networks form emergent clusters of words corresponding to syntactic (nouns, transitive and intransitive verbs) and semantic categories (e.g., animals and clothing), based solely on one child's linguistic input. The networks also acquire sensitivity to acceptability contrasts from linguistic phenomena, such as determiner-noun agreement and argument structure. We find that incorporating visual information produces an incremental gain in predicting words in context, especially for syntactic categories that are comparatively more easily grounded, such as nouns and verbs, but the underlying linguistic representations are not fundamentally altered. Our findings demonstrate which kinds of linguistic knowledge are learnable from a snapshot of a single child's real developmental experience. ---- Christian Clark Ph.D. Student Department of Linguistics The Ohio State University -------------- next part -------------- An HTML attachment was scrubbed... URL: From schuler.77 at osu.edu Wed Oct 18 19:38:09 2023 From: schuler.77 at osu.edu (Schuler, William) Date: Wed, 18 Oct 2023 23:38:09 +0000 Subject: [CaCL] Reading for 10/19 In-Reply-To: References: Message-ID: Non-paywall version: https://osf.io/preprints/psyarxiv/85k3y wm ________________________________ From: CaCL on behalf of Clark, Christian via CaCL Sent: Sunday, October 8, 2023 2:53:57 PM To: Schuler, William via CaCL Subject: [CaCL] Reading for 10/19 Hi CaCL members, CaCL will not meet on 10/12 because of autumn break. On 10/19, we will discuss "Finding Structure in One Child's Linguistic Experience" by Wang et al. (2023). Paper: https://onlinelibrary.wiley.com/doi/abs/10.1111/cogs.13305 Zoom: https://osu.zoom.us/j/93495474520?pwd=NWFRZzF6QnZrQWdheThxbWJyNjJ4dz09 Abstract: Neural network models have recently made striking progress in natural language processing, but they are typically trained on orders of magnitude more language input than children receive. What can these neural networks, which are primarily distributional learners, learn from a naturalistic subset of a single child's experience? We examine this question using a recent longitudinal dataset collected from a single child, consisting of egocentric visual data paired with text transcripts. We train both language-only and vision-and-language neural networks and analyze the linguistic knowledge they acquire. In parallel with findings from Jeffrey Elman's seminal work, the neural networks form emergent clusters of words corresponding to syntactic (nouns, transitive and intransitive verbs) and semantic categories (e.g., animals and clothing), based solely on one child's linguistic input. The networks also acquire sensitivity to acceptability contrasts from linguistic phenomena, such as determiner-noun agreement and argument structure. We find that incorporating visual information produces an incremental gain in predicting words in context, especially for syntactic categories that are comparatively more easily grounded, such as nouns and verbs, but the underlying linguistic representations are not fundamentally altered. Our findings demonstrate which kinds of linguistic knowledge are learnable from a snapshot of a single child's real developmental experience. ---- Christian Clark Ph.D. Student Department of Linguistics The Ohio State University -------------- next part -------------- An HTML attachment was scrubbed... URL: From white.1240 at osu.edu Thu Oct 19 13:19:56 2023 From: white.1240 at osu.edu (White, Michael) Date: Thu, 19 Oct 2023 17:19:56 +0000 Subject: [CaCL] Reading for 11/2 (in two weeks): Pavlick (2023) on symbols and grounding in LLMs Message-ID: Not next week but the following week: Pavlick E. 2023 Symbols and grounding in large language models. Phil. Trans. R. Soc. A 381: 20220041. https://doi.org/10.1098/rsta.2022.0041 URL to PDF: https://royalsocietypublishing.org/doi/pdf/10.1098/rsta.2022.0041 -------------- next part -------------- An HTML attachment was scrubbed... URL: From schuler.77 at osu.edu Fri Oct 27 08:55:00 2023 From: schuler.77 at osu.edu (Schuler, William) Date: Fri, 27 Oct 2023 12:55:00 +0000 Subject: [CaCL] Fwd: [Ling-Faculty] [Lingosu] Spring Course Announcement: Language Development (Psych 4554) In-Reply-To: References: Message-ID: Possibly of interest: ________________________________ From: Ling-Faculty on behalf of Wagner, Laura via Lingosu via Ling-Faculty Sent: Thursday, October 19, 2023 10:30:00 AM Subject: [Ling-Faculty] [Lingosu] Spring Course Announcement: Language Development (Psych 4554) Please pass this on to any students you know who might be interested! PSYCH 4554, LANGUAGE DEVELOPMENT This course will examine how children learn their native language, beginning with their learning of sound patterns in the womb and covering all the core components of language ? phonology, words, syntax, meaning, social uses of language, dialect, and more. We will talk about a range of approaches to studying language and how language development is influenced by inborn genetic properties as well as by children?s life experiences. This class is appropriate for students interested in Linguistics, Psychology, Education, Speech & Hearing Sciences, or anything involving language development or children. Tues & Thurs, 11:10 - 12:30, 140 Jennings Hall Instructor: Laura Wagner, Psychology [cid:436cd95b-052e-47b0-aed7-10f759ff7b8d] -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 135462 bytes Desc: image.png URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ATT00001.txt URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ATT00002.txt URL: