Kristina Gligorić

Yiwei Luo, Kristina Gligorić, Dan Jurafsky · stanford

Identifying implicit attitudes toward food can mitigate social prejudice due to food's salience as a marker of ethnic identity. Stereotypes about food are representational harms that may contribute to racialized discourse and negatively impact economic outcomes for restaurants. Understanding the presence of representational harms in online corpora in particular is important, given the increasing use of large language models (LLMs) for text generation and their tendency to reproduce attitudes in their training data. Through careful linguistic analyses, we evaluate social theories about attitudes toward immigrant cuisine in a large-scale study of framing differences in 2.1M English language Yelp reviews. Controlling for factors such as restaurant price and neighborhood racial diversity, we find that immigrant cuisines are more likely to be othered using socially constructed frames of authenticity (e.g., "authentic," "traditional"), and that non-European cuisines (e.g., Indian, Mexican) in particular are described as more exotic compared to European ones (e.g., French). We also find that non-European cuisines are more likely to be described as cheap and dirty, even after controlling for price, and even among the most expensive restaurants. Finally, we show that reviews generated by LLMs reproduce similar framing tendencies, pointing to the downstream retention of these representational harms. Our results corroborate social theories of gastronomic stereotyping, revealing racialized evaluative processes and linguistic strategies through which they manifest.

Kristina Gligorić, Tijana Zrnic, Cinoo Lee et al.

Large language models (LLMs) have shown high agreement with human raters across a variety of tasks, demonstrating potential to ease the challenges of human data collection. In computational social science (CSS), researchers are increasingly leveraging LLM annotations to complement slow and expensive human annotations. Still, guidelines for collecting and using LLM annotations, without compromising the validity of downstream conclusions, remain limited. We introduce Confidence-Driven Inference: a method that combines LLM annotations and LLM confidence indicators to strategically select which human annotations should be collected, with the goal of producing accurate statistical estimates and provably valid confidence intervals while reducing the number of human annotations needed. Our approach comes with safeguards against LLM annotations of poor quality, guaranteeing that the conclusions will be both valid and no less accurate than if we only relied on human annotations. We demonstrate the effectiveness of Confidence-Driven Inference over baselines in statistical estimation tasks across three CSS settings--text politeness, stance, and bias--reducing the needed number of human annotations by over 25% in each. Although we use CSS settings for demonstration, Confidence-Driven Inference can be used to estimate most standard quantities across a broad range of NLP problems.

Omar Shaikh, Kristina Gligorić, Ashna Khetan et al.

Effective conversation requires common ground: a shared understanding between the participants. Common ground, however, does not emerge spontaneously in conversation. Speakers and listeners work together to both identify and construct a shared basis while avoiding misunderstanding. To accomplish grounding, humans rely on a range of dialogue acts, like clarification (What do you mean?) and acknowledgment (I understand.). However, it is unclear whether large language models (LLMs) generate text that reflects human grounding. To this end, we curate a set of grounding acts and propose corresponding metrics that quantify attempted grounding. We study whether LLM generations contain grounding acts, simulating turn-taking from several dialogue datasets and comparing results to humans. We find that -- compared to humans -- LLMs generate language with less conversational grounding, instead generating text that appears to simply presume common ground. To understand the roots of the identified grounding gap, we examine the role of instruction tuning and preference optimization, finding that training on contemporary preference data leads to a reduction in generated grounding acts. Altogether, we highlight the need for more research investigating conversational grounding in human-AI interaction.

Kristina Gligoric, Tiziano Piccardi, Jake Hofman et al.

Science is facing a reproducibility crisis. Previous work has proposed incorporating data analysis replications into classrooms as a potential solution. However, despite the potential benefits, it is unclear whether this approach is feasible, and if so, what the involved stakeholders-students, educators, and scientists-should expect from it. Can students perform a data analysis replication over the course of a class? What are the costs and benefits for educators? And how can this solution help benchmark and improve the state of science? In the present study, we incorporated data analysis replications in the project component of the Applied Data Analysis course (CS-401) taught at EPFL (N=354 students). Here we report pre-registered findings based on surveys administered throughout the course. First, we demonstrate that students can replicate previously published scientific papers, most of them qualitatively and some exactly. We find discrepancies between what students expect of data analysis replications and what they experience by doing them along with changes in expectations about reproducibility, which together serve as evidence of attitude shifts to foster students' critical thinking. Second, we provide information for educators about how much overhead is needed to incorporate replications into the classroom and identify concerns that replications bring as compared to more traditional assignments. Third, we identify tangible benefits of the in-class data analysis replications for scientific communities, such as a collection of replication reports and insights about replication barriers in scientific work that should be avoided going forward. Overall, we demonstrate that incorporating replication tasks into a large data science class can increase the reproducibility of scientific work as a by-product of data science instruction, thus benefiting both science and students.

Joy Bhalla, Kristina Gligorić

Large language models exhibit sycophancy: the tendency to shift outputs toward user-expressed stances, regardless of correctness or consistency. While prior work has studied this issue and its impacts, rigorous computational linguistic metrics are needed to identify when models are being sycophantic. Here, we introduce SWAY, an unsupervised computational linguistic measure of sycophancy. We develop a counterfactual prompting mechanism to identify how much a model's agreement shifts under positive versus negative linguistic pressure, isolating framing effects from content. Applying this metric to benchmark 6 models, we find that sycophancy increases with epistemic commitment. Leveraging our metric, we introduce a counterfactual mitigation strategy teaching models to consider what the answer would be if opposite assumptions were suggested. While baseline mitigation instructing to be explicitly anti-sycophantic yields moderate reductions, and can backfire, our counterfactual CoT mitigation drives sycophancy to near zero across models, commitment levels, and clause types, while not suppressing responsiveness to genuine evidence. Overall, we contribute a metric for benchmarking sycophancy and a mitigation informed by it.

Weidi Luo, Xiaofei Wen, Tenghao Huang et al.

Large language models (LLMs) are increasingly deployed for everyday tasks, including food preparation and health-related guidance. However, food safety remains a high-stakes domain where inaccurate or misleading information can cause severe real-world harm. Despite these risks, current LLMs and safety guardrails lack rigorous alignment tailored to domain-specific food hazards. To address this gap, we introduce FoodGuardBench, the first comprehensive benchmark comprising 3,339 queries grounded in FDA guidelines, designed to evaluate the safety and robustness of LLMs. By constructing a taxonomy of food safety principles and employing representative jailbreak attacks (e.g., AutoDAN and PAP), we systematically evaluate existing LLMs and guardrails. Our evaluation results reveal three critical vulnerabilities: First, current LLMs exhibit sparse safety alignment in the food-related domain, easily succumbing to a few canonical jailbreak strategies. Second, when compromised, LLMs frequently generate actionable yet harmful instructions, inadvertently empowering malicious actors and posing tangible risks. Third, existing LLM-based guardrails systematically overlook these domain-specific threats, failing to detect a substantial volume of malicious inputs. To mitigate these vulnerabilities, we introduce FoodGuard-4B, a specialized guardrail model fine-tuned on our datasets to safeguard LLMs within food-related domains.

Myra Cheng, Kristina Gligoric, Tiziano Piccardi et al.

Anthropomorphism, or the attribution of human-like characteristics to non-human entities, has shaped conversations about the impacts and possibilities of technology. We present AnthroScore, an automatic metric of implicit anthropomorphism in language. We use a masked language model to quantify how non-human entities are implicitly framed as human by the surrounding context. We show that AnthroScore corresponds with human judgments of anthropomorphism and dimensions of anthropomorphism described in social science literature. Motivated by concerns of misleading anthropomorphism in computer science discourse, we use AnthroScore to analyze 15 years of research papers and downstream news articles. In research papers, we find that anthropomorphism has steadily increased over time, and that papers related to language models have the most anthropomorphism. Within ACL papers, temporal increases in anthropomorphism are correlated with key neural advancements. Building upon concerns of scientific misinformation in mass media, we identify higher levels of anthropomorphism in news headlines compared to the research papers they cite. Since AnthroScore is lexicon-free, it can be directly applied to a wide range of text sources.

Kristina Gligoric, Myra Cheng, Lucia Zheng et al. · stanford

The use of words to convey speaker's intent is traditionally distinguished from the `mention' of words for quoting what someone said, or pointing out properties of a word. Here we show that computationally modeling this use-mention distinction is crucial for dealing with counterspeech online. Counterspeech that refutes problematic content often mentions harmful language but is not harmful itself (e.g., calling a vaccine dangerous is not the same as expressing disapproval of someone for calling vaccines dangerous). We show that even recent language models fail at distinguishing use from mention, and that this failure propagates to two key downstream tasks: misinformation and hate speech detection, resulting in censorship of counterspeech. We introduce prompting mitigations that teach the use-mention distinction, and show they reduce these errors. Our work highlights the importance of the use-mention distinction for NLP and CSS and offers ways to address it.

Giuseppe Russo, Kristina Gligorić, Vincent Moreau et al.

Reducing meat consumption is crucial for achieving global environmental and nutritional targets. Meat-Free Day (MFD) is a widely adopted strategy to address this challenge by encouraging plant-based diets through the removal of animal-based meals. We assessed the environmental, behavioral, and nutritional impacts of MFD by implementing 67 MFDs over 18 months (once a week on a randomly chosen day) across 12 cafeterias on a large university campus, analyzing over 400,000 food purchases. MFD reduced on-campus food-related greenhouse gas (GHG) emissions on treated days by 52.9% and contributed to improved fiber (+26.9%) and cholesterol (-4.5%) consumption without altering caloric intake. These nutritional benefits were, however, accompanied by a 27.6% decrease in protein intake and a 34.2% increase in sugar consumption. Moreover, the increase in plant-based meals did not carry over to subsequent days, as evidenced by a 3.5% rebound in animal-based meal consumption on days immediately following treated days. MFD also led to a 16.8% drop in on-campus meal sales on treated days.Monte Carlo simulations suggest that if 8.7% of diners were to eat burgers off-campus on treated days, MFD's GHG savings would be fully negated. As our analysis identifies on-campus customer retention as the main challenge to MFD effectiveness, we recommend combining MFD with customer retention interventions to ensure environmental and nutritional benefits.

Stefan Krsteski, Giuseppe Russo, Serina Chang et al.

Surveys provide valuable insights into public opinion and behavior, but their execution is costly and slow. Large language models (LLMs) have been proposed as a scalable, low-cost substitute for human respondents, but their outputs are often biased and yield invalid estimates. We study the interplay between synthesis methods that use LLMs to generate survey responses and rectification methods that debias population estimates, and explore how human responses are best allocated between them. Using two panel surveys with questions on nutrition, politics, and economics, we find that synthesis alone introduces substantial bias (24-86%), whereas combining it with rectification reduces bias below 5% and increases effective sample size by up to 14%. Overall, we challenge the common practice of using all human responses for fine-tuning, showing that under a fixed budget, allocating most to rectification results in far more effective estimation.

Kaitlyn Zhou, Kristina Gligorić, Myra Cheng et al.

Although language model-based chat systems are increasingly used in daily life, most Americans remain non-adopters of chat-based LLMs -- as of June 2025, 66% had never used ChatGPT. At the same time, LLM development and evaluation rely mainly on data from adopters (e.g., logs, preference data), focusing on the needs and tasks for a limited demographic group of adopters in terms of geographic location, education, and gender. In this position paper, we argue that incorporating non-adopter perspectives is essential for developing broadly useful and capable LLMs. We contend that relying on methods that focus primarily on adopters will risk missing a range of tasks and needs prioritized by non-adopters, entrenching inequalities in who benefits from LLMs, and creating oversights in model development and evaluation. To illustrate this claim, we conduct case studies with non-adopters and show: how non-adopter needs diverge from those of current users, how non-adopter needs point us towards novel reasoning tasks, and how to systematically integrate non-adopter needs via human-centered methods.

Justyna Czestochowska, Kristina Gligoric, Maxime Peyrard et al.

Emojis come with prepacked semantics making them great candidates to create new forms of more accessible communications. Yet, little is known about how much of this emojis semantic is agreed upon by humans, outside of textual contexts. Thus, we collected a crowdsourced dataset of one-word emoji descriptions for 1,289 emojis presented to participants with no surrounding text. The emojis and their interpretations were then examined for ambiguity. We find that with 30 annotations per emoji, 16 emojis (1.2%) are completely unambiguous, whereas 55 emojis (4.3%) are so ambiguous that their descriptions are indistinguishable from randomly chosen descriptions. Most of studied emojis are spread out between the two extremes. Furthermore, investigating the ambiguity of different types of emojis, we find that an important factor is the extent to which an emoji has an embedded symbolical meaning drawn from an established code-book of symbols. We conclude by discussing design implications.

Maxime Peyrard, Beatriz Borges, Kristina Gligorić et al.

The automatic detection of humor poses a grand challenge for natural language processing. Transformer-based systems have recently achieved remarkable results on this task, but they usually (1)~were evaluated in setups where serious vs humorous texts came from entirely different sources, and (2)~focused on benchmarking performance without providing insights into how the models work. We make progress in both respects by training and analyzing transformer-based humor recognition models on a recently introduced dataset consisting of minimal pairs of aligned sentences, one serious, the other humorous. We find that, although our aligned dataset is much harder than previous datasets, transformer-based models recognize the humorous sentence in an aligned pair with high accuracy (78%). In a careful error analysis, we characterize easy vs hard instances. Finally, by analyzing attention weights, we obtain important insights into the mechanisms by which transformers recognize humor. Most remarkably, we find clear evidence that one single attention head learns to recognize the words that make a test sentence humorous, even without access to this information at training time.

Kristina Gligorić, Ryen W. White, Emre Kıcıman et al.

Nutrition is a key determinant of long-term health, and social influence has long been theorized to be a key determinant of nutrition. It has been difficult to quantify the postulated role of social influence on nutrition using traditional methods such as surveys, due to the typically small scale and short duration of studies. To overcome these limitations, we leverage a novel source of data: logs of 38 million food purchases made over an 8-year period on the Ecole Polytechnique Federale de Lausanne (EPFL) university campus, linked to anonymized individuals via the smartcards used to make on-campus purchases. In a longitudinal observational study, we ask: How is a person's food choice affected by eating with someone else whose own food choice is healthy vs. unhealthy? To estimate causal effects from the passively observed log data, we control confounds in a matched quasi-experimental design: we identify focal users who at first do not have any regular eating partners but then start eating with a fixed partner regularly, and we match focal users into comparison pairs such that paired users are nearly identical with respect to covariates measured before acquiring the partner, where the two focal users' new eating partners diverge in the healthiness of their respective food choice. A difference-in-differences analysis of the paired data yields clear evidence of social influence: focal users acquiring a healthy-eating partner change their habits significantly more toward healthy foods than focal users acquiring an unhealthy-eating partner. We further identify foods whose purchase frequency is impacted significantly by the eating partner's healthiness of food choice. Beyond the main results, the work demonstrates the utility of passively sensed food purchase logs for deriving insights, with the potential of informing the design of public health interventions and food offerings.

Kristina Gligorić, Ashton Anderson, Robert West

In November 2017, Twitter doubled the maximum allowed tweet length from 140 to 280 characters, a drastic switch on one of the world's most influential social media platforms. In the first long-term study of how the new length limit was adopted by Twitter users, we ask: Does the effect of the new length limit resemble that of the old one? Or did the doubling of the limit fundamentally change how Twitter is shaped by the limited length of posted content? By analyzing Twitter's publicly available 1% sample over a period of around 3 years, we find that, when the length limit was raised from 140 to 280 characters, the prevalence of tweets around 140 characters dropped immediately, while the prevalence of tweets around 280 characters rose steadily for about 6 months. Despite this rise, tweets approaching the length limit have been far less frequent after than before the switch. We find widely different adoption rates across languages and client-device types. The prevalence of tweets around 140 characters before the switch in a given language is strongly correlated with the prevalence of tweets around 280 characters after the switch in the same language, and very long tweets are vastly more popular on Web clients than on mobile clients. Moreover, tweets of around 280 characters after the switch are syntactically and semantically similar to tweets of around 140 characters before the switch, manifesting patterns of message squeezing in both cases. Taken together, these findings suggest that the new 280-character limit constitutes a new, less intrusive version of the old 140-character limit. The length limit remains an important factor that should be considered in all studies using Twitter data.