Papers for
ai writing tool developers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Large language models often stereotype older adults in stories
"Despite her age, Margaret was eager to make a change": Unpacking Aging Representations and Stereotypes in LLM-generated Stories
Abstract: As large language models (LLMs) increasingly generate everyday stories, their portrayals can shape expectations about aging. Prior work has largely examined age stereotypes in short responses, leaving unanswered whose lives models default to imagining and how age shapes characters' roles in complete narratives. We analyze 594 stories generated by Gemini, GPT, and Llama using text embeddings, LLM-assisted coding, and qualitative analysis. When age was unspecified, stories typically featured younger adults, with middle-aged protagonists largely absent. Older adults were often portrayed through loss, familiar routines, and help from younger characters. Even positive stories sometimes treated older adults' achievements as surprising or portrayed others making decisions on their behalf. We suggest evaluating not only the tone of generated stories, but also who appears, what opportunities they receive, and who makes decisions. Involving older adults in these evaluations can guide writing tools toward more varied, less stereotyped portrayals of later life.
Measuring human input in AI writing is unreliable using likelihood scores
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing?
Abstract: AI-text detection generally asks whether a machine wrote a text. As AI becomes an increasingly integral and accepted part of writing, the more relevant question is no longer whether AI was used, but how much of the text reflects human contribution. A prominent line of work estimates this by measuring how predictive a person's prompt is of the final text. We show that this measure can be easily manipulated. We introduce SPECIFICITY PADDING that simply asks the model to copy a few random IDs from the prompt, and that alone makes a title-only request look more human than one that supplies the article's key ideas. A context-swap audit then holds each document fixed, and only shows or hides the IDs to the evaluator. The score of the unchanged text swings by a third to half of its scale, and it rises for every document we test, on news and scientific abstracts, across varied LLMs from 3.8B to 753B parameters. The same articles can average 85\% human or 15\% human depending only on which prompt is on record, and even with no prompt at all, the output-only variant rewards the random strings themselves. An exact identity, the credit ledger, explains why. The score is a surprisal-weighted average over the text, so rare strings that the context makes copyable collect the credit, whatever they mean. A two-point bound marks a deeper limit. If the prompt could have been machine-written, every transcript-only method must, in the worst case, miss by half the contribution at stake. Likelihood-based scores alone cannot reliably establish human contribution without evidence of what the person added and how the text was produced.
GPT models often soften claims to please reviewers not reflect evidence
Writing for the Reviewer: Defensive Writing in GPT Models
Abstract: Researchers increasingly use ChatGPT to revise their papers, and recent GPT versions often narrow or even retract the authors' claims. We call such changes defensive writing when the given material does not support them, and we test two explanations: the model corrects the authors' overclaiming, or it writes for an anticipated reviewer. We ask GPT versions and models from other developers to rewrite paragraphs from papers written before ChatGPT, or to write from an evidence sheet that lists a paper's method and results. Defensive writing grows with GPT version. GPT-6-astra retracts the authors' claims outright, and when it writes from the evidence sheet, it still adds the most ungrounded qualifications. The results favor the anticipated-review explanation, and correcting overclaiming explains only a small part. When the models are only asked to polish, defense stays near the level of the originals; mentioning review raises it, and one round of self-review raises it further. At the same time, fewer than one in ten of the claims GPT-6-astra retracts are overstated. AI reviewers score defensive rewrites higher, while human readers find them harder to read and the authors less certain. Combining AI writing with AI review may amplify this style.
Antithesis use in nlp papers rises and often annoys readers
I would rather quit NLP than read another paper like this: The rise of antithesis in NLP papers
Abstract: For better or worse, LLMs are by now used routinely for scientific writing.\footnote{This paper is no exception; we did use AI to assist with writing some of the sections (see Acknowledgments).} Many have noticed that recent models fill papers with unnecessary antithesis, stating over and over what the work does not do, in ways that do not contribute to its precision or quality of expression and annoy reviewers \emph{rather than impressing them}. We study the construction \emph{rather than} in ACL papers from 2019, ACL-style arXiv papers from 2026, and papers written by GPT models from the same titles and abstracts. Its rate in 2026 is seven times the 2019 rate, and higher still in the GPT papers. Two annotators, blind to the source, find almost no 2019 use \emph{annoying} and about one in ten 2026 uses; they seldom agree on which, yet about half of 2026 papers contain a use that annoys each of them. \emph{Annoying} uses present the rejected alternative less favorably than legitimate uses. Raters of preference data and open reward models favor the construction, and an instruction to be honest promotes it. We conjecture that it is a side effect of post-training on pairwise preferences, which credit a disavowal in a single response and cannot register its cost across a text.