Excited to travel to ICLR 🇧🇷 to present our work on textual creativity metrics! See me at the 10:30am-1pm Saturday poster session in Pavilion 3 or reach out to chat 🙂

🚨Paper on AI & Copyright

Courts have credited AI companies' claims that alignment prevents reproducing copyrighted data.

What if finetuning on a simple writing task breaks it.

Worse: tuning on just one author (e.g., Murakami) unlocks verbatim recall of 30+ other authors' books (up to 90%)

(1/n)🧵

⚛️ Introducing CREATE, a benchmark for creative associative reasoning in LLMs.

Making novel, meaningful connections is key for scientific & creative works.

We objectively measure how well LLMs can do this. 🧵👇

Our paper “Inferring fine-grained migration patterns across the United States” is now out in @natcomms.nature.com! We released a new, highly granular migration dataset. 1/9

Merriam-Webster’s human editors have chosen ‘slop’ as the 2025 Word of the Year.

Day 7 of #30DayMapChallenge asked us to think about accessibility. @gsagostini.bsky.social considers two metrics of access simultaneously: distance to a Subway and distance to the subway.

We didn’t observe this negative relationship between n-gram novelty and pragmaticality in humans, only in open-source LLMs!

Death of the Novel(ty): Beyond n-Gram Novelty as a Metric for Textual Creativity N-gram novelty is widely used to evaluate language models' ability to generate text outside of their training data. More recently, it has also been adopted as a metric for measuring textual creativity...

See more details in the paper!
Work with my amazing mentors & collaborators @najoung.bsky.social, @tuhinchakr.bsky.social, Smaranda Muresan
Paper link: www.arxiv.org/abs/2509.22641
Github link: github.com/asaakyan/ngr...

On OOD dataset StyleMirror, we find that LLM-Judge novelty scores are associated with expert preferences to a larger extent than a previously proposed n-gram novelty metric, Creativity Index, suggesting our operationalization yields a more aligned metric for textual creativity.

Writing quality reward model scores are associated with both creativity and pragmaticality judgements, but are not interpretable. LLM-judge can replicate some expert novelty judgements but struggle with identifying non-pragmatic expressions.

In a follow-up study with GPT-5 and Claude, we observe that the rate of human-judged creative expressions in AI-written text is significantly lower than in human-written text.

Further, we find that both open source models tested, OLMo-1 and 2 of 7B and 32B size, exhibit a negative relationship between n-gram novelty and pragmaticality. As open-source LLMs try to generate text not present in data, their expressions tend to make less sense in context.

N-gram novelty is not a reliable metric of creativity: over *90%* of top-quartile n-gram novelty expressions were not judged as creative. We find many examples of low n-gram novelty expressions rated creative and high n-gram novelty expressions rated as non-pragmatic.

We recruit expert writers with MFA/MA/PhD background. They rated expressions in human- and AI-generated (from fully (code + DATA) open-source OLMo models) passages for if they make sense, are pragmatic, and are novel; they could also highlight any creative expressions.

The standard definition of creativity states the product has to be both novel AND appropriate. Similarly, we operationalize textual creativity as human-judged expression novelty AND sensicality (making sense by itself) + pragmaticality (making sense in context).

N-gram novelty is widely used to evaluate language models' ability to generate text outside of their training data. More recently, it has also been adopted as a metric for measuring textual creativity. However, theoretical work on creativity suggests that this approach may be inadequate, as it does not account for creativity's dual nature: novelty (how original the text is) and appropriateness (how sensical and pragmatic it is). We investigate the relationship between this notion of creativity and n-gram novelty through 7542 expert writer annotations (n=26) of novelty, pragmaticality, and sensicality via close reading of human and AI-generated text. We find that while n-gram novelty is positively associated with expert writer-judged creativity, ~91% of top-quartile expressions by n-gram novelty are not judged as creative, cautioning against relying on n-gram novelty alone. Furthermore, unlike human-written text, higher n-gram novelty in open-source LLMs correlates with lower pragmaticality. In an exploratory study with frontier close-source models, we additionally confirm that they are less likely to produce creative expressions than humans. Using our dataset, we test whether zero-shot, few-shot, and finetuned models are able to identify creative expressions (a positive aspect of writing) and non-pragmatic ones (a negative aspect). Overall, frontier LLMs exhibit performance much higher than random but leave room for improvement, especially struggling to identify non-pragmatic expressions. We further find that LLM-as-a-Judge novelty scores from the best-performing model were predictive of expert writer preferences.

N-gram novelty is widely used as a measure of creativity and generalization. But if LLMs produce highly n-gram novel expressions that don’t make sense or sound awkward, should they still be called creative? In a new paper, we investigate how n-gram novelty relates to creativity.

Are you a researcher using computational methods to understand cities?

@mfranchi.bsky.social @jennahgosciak.bsky.social and I organize an EAAMO Bridges working group on Urban Data Science and we are looking for new members!

Fill the interest form on our page: urban-data-science-eaamo.github.io

📢 New paper: Applied interpretability 🤝 MT personalization!

We steer LLM generations to mimic human translator styles on literary novels in 7 languages. 📚

SAE steering can beat few-shot prompting, leading to better personalization while maintaining quality.

🧵1/

Understanding Figurative Meaning through Explainable Visual Entailment Large Vision-Language Models (VLMs) have demonstrated strong capabilities in tasks requiring a fine-grained understanding of literal meaning in images and text, such as visual question-answering or vi...

See more experiments and details in our paper: arxiv.org/abs/2405.01474

And come see our poster at NAACL :)
Joint work by Shreyas Kulkarni, @tuhinchakr.bsky.social, Smaranda Muresan

Even powerful models achieve only 50% explanation adequacy rate, suggesting difficulties in reasoning about figurative inputs. Hallucination & unsound reasoning are the most prominent error categories.

Our main results are:
1. VLMs struggle to generalize from literal to figurative meaning understanding (training on e-ViL only achieves random F1 on our task)
2. Figurative meaning in the image is harder to explain compared to when it is in the text
3. VLMs benefit from image data during fine-tuning

Via human-AI collaboration, we augment existing datasets for multimodal metaphors, sarcasm, and humor with entailed/contradicted captions and textual explanations. The figurative part can be in the image, caption, or both. We benchmarks a variety of models on the resulting data.

We frame the multimodal figurative meaning understanding problem as an explainable visual entailment task between an image (premise) and its caption (hypothesis). The VLM predicts whether the image entails or contradicts the caption, and shows the reasoning steps in a textual explanation.

Can vision-language models understand figurative meaning in multimodal inputs, like visual metaphors, sarcastic captions or memes? Come find out at our #NAACL2025 poster on Friday at 9am!

New task & dataset of images and captions with figurative phenomena like metaphor, idiom, sarcasm, and humor.

Migration data lets us study responses to environmental disasters, social change patterns, policy impacts, etc. But public data is too coarse, obscuring these important phenomena!

We build MIGRATE: a dataset of yearly flows between 47 billion pairs of US Census Block Groups. 1/5

People often claim they know when ChatGPT wrote something, but are they as accurate as they think?

Turns out that while general population is unreliable, those who frequently use ChatGPT for writing tasks can spot even "humanized" AI-generated text with near-perfect accuracy 🎯