Morphological Inflection: A Reality Check
What it is about
The paper examines how morphological inflection benchmarks are built and scored. It names three problems: uniform sampling by type creates unnatural train-test splits, a single split gives unstable model rankings, and train and test items overlap in uncontrolled ways. The overlap is split into four types of (lemma, feature set) pairs, and the test sets are grouped as featsAttested or featsNovel. Weighted sampling gives the highest mean accuracy for Arabic with large training (77.76%), and overlap-aware sampling the lowest (61.06%). For Arabic, the mean accuracy is 77.70% on featsAttested against 41.92% on featsNovel. Models generalise poorly to new feature sets. The authors argue that in fusional languages such as English, a feature set does not map directly onto the form, so generalising to a novel set is often hard.
Ideas
- I1.
en/src/morph.rs: build an English evaluation split with novel feature sets (a feature set absent from training) and report the accuracy on those items next to the accuracy on attested items. Measure the gap on EWT test foren morph-eval; a rule-based path should show a small gap, and any large gap is a bug. - I2. Weighted test sampling in
en folds: draw the cv folds with weights by token frequency and by type, then report both. Measure the difference in the two-fold accuracy, with two identical runs first. - I3. Single-split warning: report the spread of the two cv folds next to each English score, so no ranking is claimed from one split. This makes the
cvnumbers in the release notes comparable.
For Mova
The code is en/src/morph.rs, with en folds for the two-fold check and en morph-eval for scores. The chapters are 13. Verbs and 11-nouns.md. The paper's English data is UniMorph and CHILDES material, under NC or other restrictions, so use it for tests and measurements only and never in what is built. Measure first: the attested-versus-novel gap on the English split, with the cv spread stated next to it.
Cite
@inproceedings{kodner2023morphological,
title = {{Morphological Inflection: A Reality Check}},
author = {Kodner, Jordan and Payne, Sarah and Khalifa, Salam and Liu, Zoey},
booktitle = {ACL 2023 (Toronto)},
year = {2023},
doi = {10.18653/v1/2023.acl-long.335},
eprint = {2305.15637},
archivePrefix = {arXiv},
url = {https://doi.org/10.18653/v1/2023.acl-long.335}
}References
- Javier Aguado-Orea and Julian M Pine. 2015. Comparing different models of the development of verb inflection in early child Spanish. PloS One, 10(3):e0119613. find
- Janet Bang and Aparna Nadig. 2015. Learning language in autism: Maternal linguistic input contributes to later vocabulary. Autism Research, 8(2):214–223. find
- Elizabeth Bates, Inge Bretherton, and Lynn Sebestyen Snyder. 1991. From first words to grammar: Individual differences and dissociable mechanisms, volume 20. Cambridge University Press. find
- Khuyagbaatar Batsuren, Omer Goldman, Salam Khalifa, Nizar Habash, Witold Kieraś, Gábor Bella, Brian Leonard, Garrett Nicolai, Kyle Gorman, Yustinus Ghanggo Ate, Maria Ryskina, Sabrina Mielke, Elena Budianskaya, Charbel El-Khaissi, Tiago Pimentel, Michael Gasser, William Abbott Lane, Mohit Raj, Matt… find our card
- Heike Behrens. 2006. The input–output relationship in first language acquisition. Language and cognitive processes, 21(1-3):2–24. find
- Emily M. Bender. 2009. Linguistically naïve != language independent: Why NLP needs linguistic typology. In Proceedings of the EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics: Virtuous, Vicious or Vacuous?, pages 26–32, Athens, Greece. Association for Computati… find
- MM Berl, LM Balsamo, B Xu, EN Moore, SL Wein-stein, JA Conry, PL Pearl, BC Sachs, CB Grandin, C Frattali, et al. 2005. Seizure focus affects regional language networks assessed by fMRI. Neurology, 65(10):1604–1611. find
- Johannes Bjerva and Isabelle Augenstein. 2018. From phonology to syntax: Unsupervised linguistic typology at different levels with language embeddings. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, V… find
- Damian Blasi, Antonios Anastasopoulos, and Graham Neubig. 2022. Systematic inequalities in language technology performance across the world’s languages. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5486–5505, Dublin, Irelan… find
- Lynn Bliss. 1988. The development of modals. The journal of applied developmental psychology, 9:253– 261. find
- Lois Bloom, Lois Hood, and Patsy Lightbown. 1974. Imitation in language development: If, when, and why. Cognitive psychology, 6(3):380–420. find
- Lois Masket Bloom. 1970. Language development: Form and function in emerging grammars. Ph.D. thesis, Columbia University. find
- John Neil Bohannon III and Angela Lynn Marquis. 1977. Children’s control of adult speech. Child Development, pages 1002–1008. find
- Susan R Braunwald. 1971. Mother-child communication: the function of maternal-language input. Word, 27(1-3):28–50. find
- Michael R Brent and Jeffrey Mark Siskind. 2001. The role of exposure to isolated words in early vocabulary development. Cognition, 81(2):B33–B44. find
- Giuseppe Capelli, Victoria Marrero, and María José Al-bala. 1994. Aplicación del sistema morfo a una mues-tra de lenguaje infantil. Procesamiento del Lenguaje Natural, 14. find
- Erwin Chan. 2008. Structures and distributions in morphological learning. Ph.D. thesis, University of Pennsylvania, Philadelphia, PA. find
- Jane Chandlee. 2017. Computational locality in morphological maps. Morphology, 27(4):599–641. find
- Eve V Clark. 1978. Awareness of language: Some evidence from what children say and do. In The child’s conception of language, pages 17–43. Springer. find
- Ryan Cotterell, Christo Kirov, John Sylak-Glassman, Géraldine Walther, Ekaterina Vylomova, Arya D. Mc-Carthy, Katharina Kann, Sabrina J. Mielke, Garrett Nicolai, Miikka Silfverberg, David Yarowsky, Jason Eisner, and Mans Hulden. 2018. The CoNLL– SIGMORPHON 2018 shared task: Universal morphological r… find
- Ryan Cotterell, Christo Kirov, John Sylak-Glassman, Géraldine Walther, Ekaterina Vylomova, Patrick Xia, Manaal Faruqui, Sandra Kübler, David Yarowsky, Jason Eisner, and Mans Hulden. 2017. find
- CoNLL-SIGMORPHON 2017 shared task: Universal morphological reinflection in 52 languages. In Proceedings of the CoNLL SIGMORPHON 2017 Shared Task: Universal Morphological Reinflection, pages 1–30, Vancouver. Association for Computational Linguistics. find
- Ryan Cotterell, Christo Kirov, John Sylak-Glassman, David Yarowsky, Jason Eisner, and Mans Hulden. 2016. find
- The SIGMORPHON 2016 shared Task— Morphological reinflection. In Proceedings of the 14th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 10–22, Berlin, Germany. Association for Computational Linguistics. find
- Barbara L Davis and Peter F MacNeilage. 1995. The articulatory basis of babbling. Journal of Speech, Language, and Hearing Research, 38(6):1199–1211. find
- Martha Jo-Ann Demetras. 1986. Working Parents’ Conversational Responses to their two-year-old sons. The University of Arizona. find
- Roger Brown. 1973. A first language: The early stages. Harvard University Press, Cambridge, MA. find
- Marty Demetras. 1989. Changes in parents’ conversational responses: A function of grammatical development. ASHA, St. Louis, MO. find
- Joan L Bybee. 1991. Natural morphology: The organization of paradigms and language acquisition. Crosscurrents in second language acquisition and linguistic theories, 2:67–92. find
- Marty J Demetras, Kathryn Nolan Post, and Catherine E Snow. 1986. Feedback to first language learners: The role of repetitions and clarification questions. Journal of child language, 13(2):275–292. find
- Katherine Demuth, Jennifer Culbertson, and Jennifer Alter. 2006. Word-minimality, epenthesis and coda licensing in the early acquisition of English. Language and speech, 49(2):137–173. find
- MacArthur—Bates Communicative Development Inventory in identifying language abilities of late-talking and typically developing toddlers. American Journal of Speech-Language Patholog, 14:40–51. find
- David K Dickinson and Patton O Tabors. 2001. Beginning literacy with language: Young children learning at home and school. Paul H Brookes Publishing. find
- Roy Patrick Higginson. 1985. Fixing: Assimilation in language acquisition. Ph.D. thesis, Washington State University. find
- Micha Elsner, Andrea D Sims, Alexander Erdmann, Antonio Hernandez, Evan Jaffe, Lifeng Jin, Martha Booker Johnson, Shuan Karim, David L King, Luana Lamberti Nunes, et al. 2019. Modeling morphological learning, typology, and change: What can the neural sequence-to-sequence framework contribute? Journa… find
- Christo Kirov, Ryan Cotterell, John Sylak-Glassman, Géraldine Walther, Ekaterina Vylomova, Patrick Xia, Manaal Faruqui, Sabrina J. Mielke, Arya McCarthy, Sandra Kübler, David Yarowsky, Jason Eisner, and Mans Hulden. 2018. UniMorph 2.0: Universal Morphology. In Proceedings of the Eleventh Internation… find
- Andrea Feldman. 1998. Constructing grammar: fillers, formulas, and function. Ph.D. thesis, University of Colorado at Boulder. find
- Viviana Fratini, Joana Acha, and Itziar Laka. 2014. Frequency and morphological irregularity are independent variables. Evidence from a corpus study of Spanish verbs. Corpus Linguistics and Linguistic Theory, 10(2):289–314. find
- Catherine Garvey and Robert Hogan. 1973. Social speech and social interaction: Egocentrism revisited. Child Development, pages 562–568. find
- Virginia C Gathercole. 1986. The acquisition of the present perfect: Explaining differences in the speech of Scottish and American children. Journal of Child Language, 13(3):537–560. find
- Susan A Gelman, John D Coley, Karl S Rosengren, Erin Hartman, Athina Pappas, and Frank C Keil. 1998. Beyond labeling: The role of maternal input in the acquisition of richly structured categories. Monographs of the Society for Research in Child development, pages i–157. find
- Ronald Bradley Gillam and Nils A Pearson. 2004. TNL: test of narrative language. Pro-ed Austin, TX. find
- Jean Berko Gleason. 1980. The acquisition of social speech routines and politeness formulas. In Language, pages 21–27. Elsevier. find
- Omer Goldman, David Guriel, and Reut Tsarfaty. 2022. (un)solving morphological inflection: Lemma overlap artificially inflates models’ performance. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 864–870, Dublin, Ireland. Ass… find
- William S Hall and William C Tirre. 1979. The communicative environment of young children: Social class, ethnic, and situational differences. Center for the Study of Reading Technical Report; no. 125. find
- John Heilmann, Susan Ellis Weismer, Julia Evans, and Christine Hollar. 2005. Utility of the Jordan Kodner. 2022. Computational Models of Morphological Learning. Oxford University Press. find
- Jordan Kodner, Salam Khalifa, Khuyagbaatar Batsuren, Hossep Dolatian, Ryan Cotterell, Faruk Akkus, Antonios Anastasopoulos, Taras Andrushko, Aryaman Arora, Nona Atanalov, Gábor Bella, Elena Budianskaya, Yustinus Ghanggo Ate, Omer Goldman, David Guriel, Simon Guriel, Silvia Guriel-Agiashvili, Witold… find
- Stan A Kuczaj II. 1977. The acquisition of regular and irregular past tense forms. Journal of verbal learning and verbal behavior, 16(5):589–600. find
- Constantine Lignos and Charles Yang. 2018. Morphology and language acquisition. Cambridge handbook of morphology, pages 765–791. find
- Josetxu Linaza, María Eugenia Sebastián, and Cristina del Barrio. 1981. Lenguaje, comunicación y com-prensión: Conferencia a nual de la sección de psi-cología del desarrollo de la british psychological society. Infancia y Aprendizaje, 4(sup1):195–197. find
- Zoey Liu and Emily Prud’hommeaux. 2022. Data-driven model generalizability in crosslinguistic low-resource morphological segmentation. Transactions of the Association for Computational Linguistics, 10:393–413. find
- Zoey Liu, Crystal Richardson, Richard Hatcher, and Emily Prud’hommeaux. 2022. Not always about you: Prioritizing community needs when developing endangered language technology. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3… find
- Colleen E Morisset, Kathryn E Barnard, and Cathryn L Booth. 1995. Toddlers’ language development: Sex differences within social risk. Developmental Psychology, 31(5):851. find
- Susana López Ornat. 1997. What lies in between a pre-grammatical and a grammatical representation? Evidence on nominal and verbal form-function mappings in Spanish from 1; 7 to 2; 1. Contemporary perspectives on the acquisition of Spanish, 1:3–20. find
- Katherine Nelson. 2006. Narratives from the crib. Harvard University Press. find
- Mohamed Maamouri, Ann Bies, Tim Buckwalter, and Wigdan Mekki. 2004. The Penn Arabic Treebank: Building a large-scale annotated Arabic corpus. In NEMLAR conference on Arabic language resources and tools, volume 27, pages 466–467. Cairo. find
- Brian MacWhinney. 1991. The CHILDES language project: Tools for analyzing talk. Journal of Speech, Language and Hearing Research, 40:62–74. find
- Rochelle S Newman, Meredith L Rowe, and Nan Bernstein Ratner. 2016. Input and uptake at 7 months predicts toddler vocabulary: The role of child-directed speech and infant processing skills in language development. Journal of child language, 43(5):1158– 1173. find
- Johanna G Nicholas and Ann E Geers. 1997. Communication of oral deaf and normally hearing children at 36 months of age. Journal of Speech, Language, and Hearing Research, 40(6):1314–1327. find
- Brian MacWhinney. 2000. The CHILDES Project: The Database, volume 2. Psychology Press, Abingdon-on-Thames. find
- Anat Ninio, Catherine E Snow, Barbara A Pan, and Pamela R Rollins. 1994. Classifying communicative acts in children’s interactions. Journal of communication disorders, 27(2):157–187. find
- Brian MacWhinney and Catherine Snow. 1985. The child language data exchange system. Journal of Child Language, 12(2):271–295. find
- Kemal Oflazer and Murat Saraçlar. 2018. Turkish natural language processing. Springer. find
- María del Carmen Aguirre Martínez and Sonia Mariscal Altares. 2005. Cómo adquieren los niños la gramática de su lengua: perspectivas teóricas. Editorial UNED. find
- Arya D. McCarthy, Christo Kirov, Matteo Grella, Amrit Nidhi, Patrick Xia, Kyle Gorman, Ekaterina Vylomova, Sabrina J. Mielke, Garrett Nicolai, Miikka Silfverberg, Timofey Arkhangelskiy, Nataly Krizhanovsky, Andrew Krizhanovsky, Elena Klyachko, Alexey Sorokin, John Mansfield, Valts Ernštreits, Yuval… find
- Arya D. McCarthy, Ekaterina Vylomova, Shijie Wu, Chaitanya Malaviya, Lawrence Wolf-Sonkin, Garrett Nicolai, Christo Kirov, Miikka Silfverberg, Sabrina J. Mielke, Jeffrey Heinz, Ryan Cotterell, and Mans Hulden. 2019. find
- The SIGMORPHON 2019 shared task: Morphological analysis in context and cross-lingual transfer for inflection. In Proceedings of the 16th Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 229–244, Florence, Italy. Association for Computational Linguistics. find
- Lorraine McCune. 1995. A normative study of representational play in the transition to language. Developmental psychology, 31(2):198. find
- Rosa Montes. 1987. Secuencias de clarificación en conversaciones con niños (morphe 3-4). Universidad Autónoma de Puebla. find
- Ann M Peters. 1987. The role of imitation in the developing syntax of a blind child. Text-Interdisciplinary Journal for the Study of Discourse, 7(3):289–309. find
- Tiago Pimentel, Maria Ryskina, Sabrina J. Mielke, Shijie Wu, Eleanor Chodroff, Brian Leonard, Garrett Nicolai, Yustinus Ghanggo Ate, Salam Khalifa, Nizar Habash, Charbel El-Khaissi, Omer Goldman, Michael Gasser, William Lane, Matt Coler, Arturo Oncevay, Jaime Rafael Montoya Samame, Gema Celeste Silv… find
- SIGMORPHON 2021 shared task on morphological reinflection: Generalization across languages. In Proceedings of the 18th SIGMOR-PHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 229–259, Online. Association for Computational Linguistics. find
- Steven Pinker and Michael T Ullman. 2002. The past and future of the past tense. Trends in Cognitive Sciences, 6(11):456–463. find
- V Remedi. 2014. Creación de corpus de datos sobre estudio longitudinal de adquisición de lenguaje de una niña de la región central de Argentina. Ph.D. thesis, National University of Córdoba. find
- Brian Roark and Richard Sproat. 2007. Computational approaches to morphology and syntax, volume 4. Oxford University Press. find
- Amye Warren-Leubecker. 1982. Sex differences in speech to children. Ph.D. thesis, Georgia Institute of Technology. find
- Pamela Rosenthal Rollins. 2003. Caregivers’ contingent comments to 9-month-old infants: Relationships with later language. Applied Psycholinguistics, 24(2):221– 234. find
- Silvan Wehrli, Simon Clematide, and Peter Makarov. 2022. CLUZH at SIGMORPHON 2022 shared tasks on morpheme segmentation and inflection generation. In Proceedings of the 19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 212–219, Seattle, Washington. Ass… find
- Jacqueline Sachs and KE Nelson. 1983. Talking about the there and then: The emergence of displaced reference in parent-child discourse. Children’s Language, 4:1–28. find
- Richard M Weist, Aleksandra Pawlak, and Karen Hoffman. 2009. Finiteness systems and lexical aspect in child Polish and English. Linguistics, 47(6):1321– 1350. find
- R Keith Sawyer. 2013. Pretend play as improvisation: Conversation in the preschool classroom. Psychology Press. find
- Wikimedia. 2022. Wikimedia Downloads. find
- Mark S. Seidenberg and D. Plaut. 2014. Quasiregular-ity and its discontents: The legacy of the past tense debate. Cognitive Science, 38 6:1190–228. find
- Melanie Soderstrom, Megan Blossom, Rina Foygel, and James L Morgan. 2008. Acoustical cues and grammatical units in speech to two preverbal infants. Journal of Child Language, 35(4):869–902. find
- Richard A Sprott. 1992. Children’s use of discourse markers in disputes: Form-function relations and discourse in child language. Discourse Processes, 15(4):423–439. find
- Patrick Suppes. 1974. The semantics of children’s language. American Psychologist, 29(2):103. find
- Virginia Valian. 1991. Syntactic subjects in the early speech of American and Italian children. Cognition, 40(1-2):21–81. find
- Lori J. van Houton. 1986. The role of maternal input in the acquisition process: The communicative strategies of adolescent and older mothers with the language learning children. In The Proceedings of the Boston University Conference on Language Development. find
- Ekaterina Vylomova, Jennifer White, Elizabeth Salesky, Sabrina J. Mielke, Shijie Wu, Edoardo Maria Ponti, Rowan Hall Maudslay, Ran Zmigrod, Josef Valvoda, Svetlana Toldova, Francis Tyers, Elena Klyachko, Ilya Yegorov, Natalia Krizhanovsky, Paula Czarnowska, Irene Nikkarinen, Andrew Krizhanovsky, Tia… find
- SIGMORPHON 2020 shared task 0: Typologically diverse morphological inflection. In Proceedings of the 17th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 1–39, Online. Association for Computational Linguistics. find
- Shijie Wu, Ryan Cotterell, and Mans Hulden. 2021. Applying the transformer to character-level transduction. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1901–1907, Online. Association for Computational Linguistics. find
- Shijie Wu, Ryan Cotterell, and Timothy O’Donnell. 2019. Morphological irregularity correlates with frequency. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5117–5126, Florence, Italy. Association for Computational Linguistics. find
- Karina Hess Zimmermann. 2003. El desarrollo linguístico en los años escolares: análisis de narraciones infantiles. Ph.D. thesis, El Colegio de México. find
This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/lex-2023-kodner-inflection-reality-check/