“If we spoke as we write, we should find no one to listen.
And if we wrote as we speak, we would find no one to read.“ (T.S.Eliot)
"AI won't replace humans, but humans who use AI will replace humans that don't."
(Karim Lakhani, professor at Harvard Business School)
Communication is a vital skill, and writing in particular, as it allows us to overcome the limits of space and time. Hence, writing matters, alas this is not an easy task. It has to be learned which requires time, effort and good feedback which may not be available when needed. In sum, acquiring the skill of writing can be a daunting task, presenting a challenge not only for kids learning a foreign language, but also for accomplished scientists writing in their mother tongue. There are many reasons why writing is difficult: lack of relevant feedback; down sliding or forgetting due to multitasking (time-sharing), ... Also, writing (composing) is complex. It is not just a single task, say, spelling. It is the ability to carry out different tasks, organize them into a plan modifiable at any time. For example, given a goal an author has to
analyze the problem, the context, and the audience,
determine the content (messages) and its order (outline planning),
choose the appropriate linguistic means (lexicalization, syntax, morphology),
control emphasis and discourse flow (cohesion),
determine layout,
check spelling and finally,
revise, i.e., consider rewriting at the various levels (content, form, spelling, tone), [1].
Any of these tasks may pose a problem, though at different points in time: before, during, or after the physical act of writing. Realizing the importance of language production, experts with different backgrounds (engineering, linguistics, psychology, education) have tried to build applications, to reveal the mental operations behind it, or to help people to acquire the skill of writing, i.e., composition.
The engineering approach is best represented by Large Language Models (LLMs) [2] and the book by (Reiter & Dale, 2000) [3], until recently, the bible of natural language generation (NLG). While presenting an architecture (see below) used by most developers in the field of NLG, one may nevertheless wonder whether or how well it is adapted for writing. Consider the following: in nearly all systems based on this framework, generation is performed by machines only and not with the help of humans who provide input (message, goals) or feedback (revision). Second, essential components of writing (problem-solving, brainstorming, evaluation, revision) are not part of the model. The fact that despite the large number of references, none of them refers to scholars (psychologists, expert writers) having studied it, speaks for itself [4]. Large Language Models (LLMs) have taken the world by storm. By analyzing patterns and relationships within text corpora, LLMs develop the ability to generate new text that is both coherent and contextually relevant. Given the power and efficiency of LLMs one may ask whether we still need humans to write texts. The answer to this question, and the relevance of LLMs for our goal (interactive writing) could be one of the topics of the workshop. So could be the definition of a shared task, which could be discussed in a special session (brainstorming devoted to the specifications of the task), or be one of the possible topics of submissions.
The psycholinguistic or cognitive science approach. Cognitive psychologists have used various techniques to study some of the mental activities taking place during the process of writing: verbal protocols, neuroimaging, keystroke logging, eye tracking, pausing, etc., [5]. Some of the results have been used to build models of writing (Bereiter, 1980; Bereiter & Scardamalia, 1987; van Wijk, 1999, Hayes, 1996), and while there have been even other proposals the one by Flower & Hayes (1980, 1981, 2016) has become the standard in psychology, and this is still the case despite all these years. Alas, despite its popularity among psychologists, this model does not fit our purpose. Being basically just a diagram of linked empty boxes, the model lacks too many details to be used by engineers to build an authoring aid.
The educational approach offered by writing experts. The list of books with advice for students wishing to acquire the skill of writing goes in the hundreds. Here are only a few (Kane, 1983; Lawrence, 1977; McCrimmon, 1976; Payne, 1965; Reid, 1982; Rico, 1983; Scardamalia, 1981; van Nostrand et al., 1978). For a longer list, see (Zock, 1986). There is also a huge literature with work coming from rhetoricians or psychologists interested in writing or the acquisition of the skill [6]. Alas, while very useful for a human trying to learn to write, they are not precise enough for engineers.
Given the fact that all three communities work on a very similar problem, text production, one may wonder to what extent their proposals are comparable, or compatible, i.e., completing or reconciling each other. This is one of the questions that could be addressed during the workshop. The table here below summarizes the major steps of writing or text production proposed by (a) engineers (Reiter & Dale, 2000), (b) psychologists (Flower & Hayes), and (c) experts of writing (teachers, educators).
A 1° Text planner ⇢ [text plan] ⇢ 2° Sentence planner ⇢ [sentence plan] ⇢ 3° Realizer ⇢ [sentence]
B 1° Planning ⇢ [idea generation + structuring] ⇢ 2° Translating ⇢ [linguistic form] ⇢ 3° Revising
C 1° Generate ideas ⇢ 2° Organize/plan ⇢ 3° Draft ⇢ 4° Revise ⇢ 5° Edit
As one can see, they all assume more or less a pipeline. Information flows basically only in one direction, going from top to bottom, and, if the model includes revision, it is generally delayed to the very end, while it could be useful at any level and at any moment. Writing in a natural setting is a cyclic process with many (often unpredictable) interactions regardless of the level. Hence, what is needed is an architecture flexible enough to allow for information flow in all directions like in a 'complete graph' or in a society of mind (Minsky, 1988), where all modules are directly connected and where any component may be at the centre of attention anytime (de Smedt, Horacek Zock, 1996).
Goals of the Workshop
While many workshops have been devoted to the building of educational software, Intelligent and Interactive Writing Assistants, or the detection of suboptimal output, they were mostly concerned with the lower end of writing, for example, spotting grammatical errors or spelling mistakes (Jourdan et al. 2023; Dale & Viethen, 2021; Leacock et al. 2022; Heidorn, 2000; Dale & Kilgarriff, 2010). Yet one could also include the higher levels, and check the quality of the facts, their relevancy and dorganization, or the soundness of reasoning. Our proposal seems to be the first of its kind (a) to consider the entire spectrum of writing (planning: ideation, organization; drafting/expression; editing; revision); (b) to include the higher levels of composition, (c) to integrate right from the start humans into the loop of the development cycle, and (d) to consider support and feedback at any moment (before, during, and after writing) rather than only at the very end.
Both humans and machines can produce text, but they do so in very different ways (Campbell, 2023; Mahlow, 2023a, 2023b; Bazerman, 2018) and at different speeds. Each method has its qualities and shortcomings. Hence the question, how to build a bridge between the two to get the best of both worlds? Writing is a difficult, time-consuming task, yet very important. This being so one may wonder when and how machines can help (Strobl et al., 2019)? Our goal is not writing at the speed of thought, but rather writing that reflects the steps or results of our thinking. Being practically minded, we do not care how machines do their job as long as they can help. Hence, the main question is not 'shall we use GPT to produce texts,' but 'how and when (what level)' to use LLMs (or other techniques) to help people to produce written text?
Footnotes
As one can see, writing is more than just linguistic knowledge. Next to domain expertise it implies also higher order cognitive skills like thinking, reasoning and strategic planning (Zock & Tesfaye, 2017).
For details see the book by (Narayan & Gardent, 2022), or the surveys by (Becker et al. 2024; Zhang et al. 2023; Lin et al., 2023; Noureen et al., 2022, Iqbal et al. 2022). For criticisms, see (Monett & Paquet, 2025; Marcus, 2022; Yu, et al., 2022; Bender et al. 2021; Dale, 2020).
For more recent work see the surveys by (Gatt & Krahmer, 2018), or (Bateman & Zock, 2022). Also, here are two curated lists referring to work on 'composition/writing' and ‘language production’.
For example, none of the following names appears in the book : de Beaugrande, 1984; Flower, 1989; Hayes, 1996; van Dijk, 1980; van Dijk & Kintsch, 1983; Scardamalia, 1981; Kellog, 1999; Elbow, 1998; Emig, 1983; Sharples et al., 1992, despite the fact that all of them have greatly contributed to our understanding of writing and the process.
Sullivan & Lindgren, 2021; Gregg & Steinberg, 2016; Leijten & Van Waes, 2013; Wengelin, et al., 2009; Galbraith, et al. 2007; Andriessen et al., 1996; Ericsson & Simon, 1993; Caccamise, 1987; Hayes & Flower, 1981.
Kruse et al. 2023; Horowitz, 2023; Lindgren et al., 2019; Tynjälä, et al. 2012; Fayol et al. 2012; Bazerman, 2009; MacArthur et al., 2008; Sanders et al. 2006; Allal et al. 2004; Nystrand, 2003; Alamargot & Chanquoy, 2001; McCutchen, 2000; Elbow, 1998; Flower (1989); de Beaugrande, 1984; Matsuhashi et al. 1987; Nystrand, 1982; Black, J., et al. 1982; Bruce et al. 1982; Martlew, 1982; Collins. & Gentner, 1980; Gregg & Steinberg, 1980; Mayer, 1979; Mayer, 1978; Freedle, 1977; Meyer, 1975; Emig, 1971. Horowitz (2023) and Flower (1989) being two excellent starting points.
References
Alamargot, D., & Chanquoy, L. (2001). Through the models of writing. Dordrecht-Boston. London: Kluwer Academic Publishers.
Allal, L., Chanquoy, L. & Largy, P. (Eds.) (2004). Revision: cognitive and instructional processes. Kluwer.
Andriessen J., deSmedt K., & Zock, M. (1996). Discourse Planning: Empirical Research and Computer Models. In T. Dijkstra & K. de Smedt (Eds), Computational Psycholinguistics: AI and Connectionist Models of Human Language Processing, Taylor & Francis, London, pp. 247–278.
Bateman, J., & Zock, M. (2022). Natural language generation. In R. Mitkov (Ed. ), Handbook of Computational Linguistics, London: Oxford University Press, pp. 747–769.
Bazerman, C. (2009). Handbook of research on writing: history, society, school, individuals, text. Routledge.
Bazerman, C. (2018). What do humans do best? Developing communicative humans in the changing socio-cyborgian landscape. In S. Logan & W. Slater (Eds.), Perspectives on Academic and Professional Writing in an Age of Accountability (pp. 187–203). Southern Illinois University Press.
Becker, J., et al. (2024). Text Generation: A Systematic Literature Review of Tasks, Evaluation, and Challenges." arXiv preprint arXiv:2405.15604 (2024).
Bender, E. M., Gebru, T., McMillan-Major, A. & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? Proceedings of the 2021 ACM conference on fairness, accountability, and transparency (pp. 610–623).
Bereiter, C. & Scardamalia, M. (1987). The psychology of written composition. Hillsdale, NJ: Lawrence Erlbaum Associates. (2013) Routledge.
Black, J., Wilkes-Gibbs, D. & Gibbs, R. (1982). What writers need to know that they don't know they need to know, in: Nystrand, M. Bruce, B., Collins, A., Rubin, A. & Gentner, D. (1982). Three perspectives on writing, Educational Psychologist, vol. 17, n° 3.
Caccamise, D. J. (1987). Idea generation in writing. In A. Matsuhashi (Ed.), Writing in Real Time? (pp. 224–253) Norwood, NJ: JAI Press Limited.
Campbell, S. H. (2023). What is human about writing? Writing process theory and ChatGPT.
Chung, J., Kamar, E. & Amershi, S. (2023). Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions. arXiv preprint arXiv:2306.04140.
Coch, J. (1996). Evaluating and comparing three text-production techniques. In COLING, The 16th International Conference on Computational Linguistics.
Collins, A. M. & Gentner, D. (2016/1980). A framework for a cognitive model of writing. In L. W. Gregg & E. R. Steinberg (Eds.), pp. 51–72.
Dale, R. (2020). Natural language generation: The commercial state of the art in 2020. Natural Language Engineering, 26(4), 481-487.
Dale, R. & Kilgarriff, A. (2010). Helping Our Own: Text massaging for computational linguistics as a new shared task. In Proceedings of the 6th International Natural Language Generation Conference.
Dale, R., & Viethen, J. (2021). The automated writing assistance landscape in 2021. Natural Language Engineering,27(4), 511-518.
De Beaugrande, R. (1984). Text production: Toward a science of composition (Vol. 11). Ablex Publishing Corporation.
De Smedt, K., H. Horacek & M. Zock (1996) Architectures for Natural Language Generation: Problems and Perspectives. In G. Adorni & M. Zock (Eds.), Trends in Natural Language Generation: an Artificial Intelligence Perspective, Springer Verlag, New York, pp. 17-46.
Elbow, P. (1998). Writing with power: Techniques for mastering the writing process. Oxford University Press.
Ellouze, N., Ben Ahmed, M. & Métais, E. (2008). State-of-the-art topic map-building approaches. In International Workshop on Model-Based Software and Data Integration (pp. 102-112). Berlin, Heidelberg: Springer Berlin Heidelberg.
Emig, J. (1983). The web of meaning: Essays on writing, teaching, learning, and thinking. Upper Montclair, NJ: Boynton.
Ericsson, K.A. & Simon, H.A. (1993). Protocol analysis: Verbal reports as data (2d ed.). Cambridge, MA: MIT Press.
Fayol, M., Alamargot, M. D., & Berninger, V. (Eds.). (2012). Translation of thought to written text while composing: Advancing theory, knowledge, research methods, tools, and applications. Psychology Press.
Flower, L. (1989). Problem-solving strategies for writing. Orders, Harcourt Brace Jovanovich, 7555 Caldwell Ave., Chicago, IL
Flower, L. S. & Hayes, J. R. (2016). In Gregg & Steinberg (Eds.).The dynamics of composing: Making plans and juggling constraints. In Cognitive processes in writing (pp. 31-50). Routledge.
Flower, L. & Hayes, J. R. (1980). The dynamics of composing: Making plans and juggling constraints. In L. Gregg and E. Steinberg (Eds.), Cognitive processes in writing: An interdisciplinary approach. Hillsdale, NJ: Lawrence Erlbaum.
Flower, L. & Hayes, J. R. (1981). A cognitive process theory of writing. College composition and communication,32(4), 365–387.
Freedle, R. (ed) (1977). Discourse production and comprehension, Ablex, Norwood, New Jersey.
Galbraith, D., Torrance, M. & Waes, L. (Eds.). (2007). Writing and cognition: Research and applications. Elsevier.
Gatt, A. & E. Krahmer (2018). ‘Survey of the State of the Art in Natural Language Generation: Core Tasks, Applications and Evaluation.’ (https://www.jair.org/index.php/jair/article/view/11173). Journal of Artificial Intelligence Research 61: 65–170.
Goldberg, E., Driedger, N. & Kittredge, R. I. (1994). Using natural-language processing to produce weather forecasts. IEEE Expert, 9(2), 45–53.
Gregg, L. W. & Steinberg, E. R. (Eds.). (2016). Cognitive processes in writing. Routledge.
Hayes, J. R. & Flower, L. (1981). Uncovering cognitive processes in writing: An introduction to protocol analysis.ERIC Clearinghouse.
Hayes, J.R. (1996). A new framework for understanding cognition and affect in writing. In C. M. Levy & S. Ransdell (Eds.). The science of writing: Theories, methods, individual differences and applications (pp. 1–27). Mahwah, NJ: Lawrence Erlbaum Associates.
Hayes, J.R. & Flower, L.S. (1980). Identifying the organization of writing processes. In L. W. Gregg & E.R. Steinberg (Eds.), Cognitive processes in writing: An interdisciplinary approach (pp. 3–30). Hillsdale, NJ: Lawrence Erlbaum.
Heidorn, G. (2000). Intelligent writing assistance. In R. Dale, H. Moisl, and H. Somers, editors, Handbook of Natural Language Processing, pages 181–207. Marcel Dekker, Inc.
Horowitz, R. (Ed.). (2023). The Routledge international handbook of research on writing. Taylor & Francis.
Iqbal, T. & Shaima Q. (2022). The survey: Text generation models in deep learning." Journal of King Saud University- Computer and Information Sciences 34.6 (2022): 2515-2528.
Jourdan, L., Boudin, F., Dufour, R. & Hernandez, N. (2023). Text revision in scientific writing assistance: An overview. arXiv preprint arXiv:2303.16726.
Kane, T. (1983). The Oxford Guide to Writing, Oxford University Press, Oxford.
Kellogg, R. T. (1999). The psychology of writing. Oxford University Press.
Kintsch, W. (1987). Psychological processes in discourse production. In H.W. Dechert & M. Raupach (Eds), Psycholinguistic models of production (pp. 163-180). Norwood, N.J.: Ablex.
Kintsch, W. & van Dijk, T.A. (1978). Toward a model of text comprehension and production. Psychological Review, 85, 363-394.
Kruse, O., Rapp, C., Anson, C. M., Benetos, K., Cotos, E., Devitt, A., & Shibani, A. (2023). Digital Writing Technologies in Higher Education: Theory, Research, and Practice.
Lawrence, M. (1977). Writing as a thinking process, The University of Michigan Press, Ann Arbor.
Leacock, C., Gamon, M., Mejia, J. A. & Chodorow, M. (2022). Automated grammatical error detection for language learners. Springer Nature.
Leijten, M. & Van Waes, L. (2013). Keystroke logging in writing research: Using Inputlog to analyze and visualize writing processes. Written Communication, 30(3), 358-392.
Lin, Y, et al. (2023). A survey on neural data-to-text generation. IEEE Transactions on Knowledge and Data Engineering.
Lindgren, E. & Sullivan, K. (Eds.). (2019). Observing writing: Insights from keystroke logging and handwriting. Brill.
MacArthur, C.A. (2008). Writing Process from a Cognitive Perspective. In MacArthur, C. A., Graham, S. & Fitzgerald, J. (Eds.). Handbook of writing research. Guilford Press.
Mahlow, C. (2023a). Large Language Models and Artificial Intelligence as Tools for Teaching and Learning Writing. Osnabrücker Beiträge zur Sprachtheorie, 101, 175-196.
Mahlow, C. (2023b). Writing tools: Looking back to look ahead. arXiv preprint arXiv:2303.17894. Martlew, M. (ed) (1982). The psychology of writing: A developmental approach, London, Wiley.
Marcus, G. (2022). Deep learning is hitting a wall. Nautilus. https://nautil.us/deep-learning-is-hitting-a-wall-238440/
Matsuhashi, A. (ed) (1987). Writing in real time: Modeling production processes, London, Longmans.
Mayer, R. (1978). Advance organizers that compensate for the organization of text, J.of Educational Psychology.
Mayer, R. (1979). Can advance organizers influence meaningful learning, Review of Educational Research, 49.
McCrimmon, J. (1976). Writing with a purpose, Boston, Houghton Mifflin.
McCutchen, D. (2000). Knowledge, Processing and Working Memory: implications for the theory of writing. Educational Psychologist,35(1), 13-23.
McKeown, K., Kukich, K. & Shaw, J. (1994). Practical issues in automatic documentation generation. Practical issues in automatic document generation. In Proceedings of the Fourth Conference on Applied Natural-Language Processing (ANLP-I994), 7-14.
Mellish, C., Scott, D., Cahill, L., Paiva, D., Evans, R. & Reape, M. (2006). A reference architecture for natural language generation systems. Natural language engineering, 12(1), 1-34.
Meyer, B. J. F. (1975). The organization of prose and its effects on memory (Vol. 1). North Holland.
Minsky, M. (1988). Society of mind. Simon and Schuster.
Monett, D. & Paquet, G. (2025). The commodification of education and the (generative) AI-induced scam-like culture. https://www.researchgate.net/publication/383148782
Mosenthal, P. & Tamor, L. & Walmsley, S. (1983). Research on writing, Longman, New York
Narayan, S. & Gardent, C. (2020). Deep Learning Approaches to Text Production. Synthesis Lectures on Human Language Technologies, 13(1), 1-199.
Narayan, S., & Gardent, C. (2022). Deep learning approaches to text production. Springer Nature.
Navigli, R. & Ponzetto, S. P. (2012). BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artificial intelligence, 193, 217-250.
Niquet, G. (1978). Structurer sa pensée, structurer sa phrase (Organize your thoughts, structure your sentences), Hachette, Paris.
Norman, D. A. (1989). Cognitive artifacts. In J. M. Carroll (Ed.), Designing interaction: Psychology at the human-computer interface (pp. 17-38). New York: Cambridge University Press.
Noureen, N., et al. (2022). A systematic literature review on text generation using deep neural network models." IEEE Access 10 (2022): 53490-53503.
Nystrand M. (ed.) (1982). What writers know: The language, process, and structure of written discourse. New York: Academic Press.
Nystrand, M. & Duffy, J. (Eds.). (2003). Towards a rhetoric of everyday life: New directions in research on writing, text, and discourse.
Paris, C., Vander Linden, K., Fischer, M., Hartley, A., Pemberton, L., Power, R. & Scott, D. (1995). A support tool for writing multilingual instructions. In International Joint Conference on Artificial Intelligence (Vol. 14, pp. 1398-1404). Lawrence Erlbaum.
Pascali, A. (2023). Guided text generation with Large Language Models. Medium, https://medium.com/productizing-language-models/guided-text-generation-with-large-language-models-d88fc3dcf4c
Payne, L.V. (1965). The lively art of writing, Mentor books, New York.
Reid, J. (1982). The Process of composition, Prentice Hall, New Jersey.
Reiter, E. & Dale, R. (2000). Building Natural Language Generation Systems. Cambridge University Press.
Rico, G. (1983). Writing the natural way, Tarcher, Los Angeles
Roget, P. (1852). Thesaurus of English Words and Phrases : classified and arranged so as to facilitate the expression of ideas and assist in literary composition Longman, London.
Sanders, T. J. & Schilperoord, J. (2006). Text structure as a window on the cognition of writing. In C. MacArthur, S. Graham, y Fitzgerald, J. (Eds.), Handbook of writing research (pp. 357-402). New York: The Guilford Press.
Sanders, T. & Van Wijk, C. (2013). Text analysis as a research tool: How hierarchical text structure contributes to the understanding of conceptual processes in writing. In Levy, C. M. & Ransdell, S. (Eds.). The science of writing (pp. 251-269). Routledge.
Scardamalia, M. (1981). Writing for Results: A Sourcebook of Consequential Composing Activities. The Ontario Institute for Studies in Education, 252 Bloor St. West, Toronto, Ontario M5S 1V6 Canada.
Scardamalia, M. & Bereiter, C. (1984). Development of strategies in text processing. In Mandl, H., Stein, N. & T. Trabasso (eds.), Learning and comprehension of text (pp. 379-406).
Sharples, M., Goodlet, J. & Pemberton, L. (1992). Developing a writer's assistant In J. Hartley (Ed.), Technology and writing: Readings in the psychology of written communication (pp. 209—220). London: Jessica Kingsley Publishers.
Strobl, C., Ailhaud, E., Benetos, K., Devitt, A., Kruse, O., Proske, A., & Rapp, C. (2019). Digital support for academic writing: A review of technologies and pedagogies. Computers & education, 131, 33-48.
Sullivan, K. & Lindgren, E. (Eds.). (2021). Computer keystroke logging and writing. Brill.
Tynjälä, P., Mason, L. & Lonka, K. (Eds.). (2012). Writing as a learning tool: Integrating theory and practice (vol. 7). Springer Science &Business Media.
van Dijk, T. (1980). Semantic macrostructures, Lawrence Erlbaum, New York.
van Dijk, T. & Kintsch,W. (1983). Strategies of discourse comprehension, Academic Press,New York .
van Nostrand et al. (1978). Functional writing, Boston, Houghton - Mifflin.
van Wijk, C. & T. Sanders (1999). Identifying writing strategies through text analysis. Written Communication, 16, 51-75
Yu, W. et al. (2022). A survey of knowledge-enhanced text generation." ACM Computing Surveys 54.11s (2022): 1-38.
Zhang, H., et al. (2023). A survey of controllable text generation using transformer-based pre-trained language models." ACM Computing Surveys, 56.3, pp. 1-37.
Zock, M. & D. Tesfaye (2017). Use your mind and learn to write: the problem of producing coherent text. In Sharp, B., Sedes, F. & Lubaszewski, W. (Eds.). Cognitive Approaches to Natural Language Processing. ISTE, London, pp. 129-158.
Zock, M. (1986). Le fil d’Ariane ou les grammaires de texte comme guide dans l’organisation et l’expression de la pensée en langue maternelle et/ou étrangère. (Ariadne's thread or text grammars as a guide in the organization and expression of thought in the mother tongue or in a foreign language), Paris: UNESCO.