Vasudha Varadarajan

Andrew Aquilina, Chetna Nihalani, Vasudha Varadarajan et al.

LLM chatbots increasingly serve as a first source of support for people in psychological distress, including those whose distress is entangled with delusional beliefs. Prior work on LLM mental-health safety largely evaluates general therapeutic quality or single-turn crisis detection, leaving unclear how models behave when distress is intertwined with delusion over sustained conversations. We address this gap with matched multi-turn simulations, across clinically grounded personas and six LLMs, that pair each delusional conversation with a distress-only control to isolate the effect of delusional framing. This reveals a recognition-intervention gap: models detect distress at comparable rates regardless of framing, yet sharply fail to act on it once distress is embedded in delusion, with safety interventions suppressed by up to 4.5x. The failure tracks accumulated acceptance of the user's premises rather than emotional validation. Worse, the intuitive fix of prompting models to assess user distress backfires under delusional framing; only delusion-aware prompting with explicit response guidance closes the gap, and even this depends on a delusion classifier that is itself unreliable on the most vulnerable models. Safe deployment therefore requires treating delusional framing as a distinct risk signal that overrides conversational accommodation.

Leena Mathur, Jenny T. Liang, Vasudha Varadarajan et al.

AI researchers have been advancing socially intelligent AI agents (Social-AI) across embodiments, from chatbots to physical robots. As Social-AI is increasingly deployed in everyday settings, decisions about the roles these agents should play will depend on how laypeople perceive them. However, public perceptions of social intelligence in AI agents and the acceptability of these agents remain largely understudied. We present a mixed-methods survey of adults in the United States (N=200) that examines social intelligence as a perceived construct in AI agents. Our survey investigates the extent to which participants believe current AI agents have social intelligence, abilities of agents that participants associate with social intelligence, contextual factors influencing participant acceptance of Social-AI agents, and concerns participants hold about these technologies. Participants widely reported having already encountered AI agents they perceived as socially intelligent and grounded their judgments in observable behaviors, more than beliefs about AI agency or intent. We identified a support-adoption gap in acceptability judgments: participants supported the existence of Social-AI agents for others far more than for their own personal use. Our analysis uncovers layperson concerns about Social-AI, informing AI governance regarding appropriate deployment contexts, agent roles, and risks to end users.

Vasudha Varadarajan, Sverker Sikström, Oscar N. E. Kjell et al.

Mental health issues differ widely among individuals, with varied signs and symptoms. Recently, language-based assessments have shown promise in capturing this diversity, but they require a substantial sample of words per person for accuracy. This work introduces the task of Adaptive Language-Based Assessment ALBA, which involves adaptively ordering questions while also scoring an individual's latent psychological trait using limited language responses to previous questions. To this end, we develop adaptive testing methods under two psychometric measurement theories: Classical Test Theory and Item Response Theory. We empirically evaluate ordering and scoring strategies, organizing into two new methods: a semi-supervised item response theory-based method ALIRT and a supervised Actor-Critic model. While we found both methods to improve over non-adaptive baselines, We found ALIRT to be the most accurate and scalable, achieving the highest accuracy with fewer questions (e.g., Pearson r ~ 0.93 after only 3 questions as compared to typically needing at least 7 questions). In general, adaptive language-based assessments of depression and anxiety were able to utilize a smaller sample of language without compromising validity or large computational costs.

Eunkyu Park, Wesley Hanwen Deng, Vasudha Varadarajan et al.

Explanations are often promoted as tools for transparency, but they can also foster confirmation bias; users may assume reasoning is correct whenever outputs appear acceptable. We study this double-edged role of Chain-of-Thought (CoT) explanations in multimodal moral scenarios by systematically perturbing reasoning chains and manipulating delivery tones. Specifically, we analyze reasoning errors in vision language models (VLMs) and how they impact user trust and the ability to detect errors. Our findings reveal two key effects: (1) users often equate trust with outcome agreement, sustaining reliance even when reasoning is flawed, and (2) the confident tone suppresses error detection while maintaining reliance, showing that delivery styles can override correctness. These results highlight how CoT explanations can simultaneously clarify and mislead, underscoring the need for NLP systems to provide explanations that encourage scrutiny and critical thinking rather than blind trust. All code will be released publicly.

Nikita Soni, August Håkan Nilsson, Syeda Mahwish et al.

Mental health is not a fixed trait but a dynamic process shaped by the interplay between individual dispositions and situational contexts. Building on interactionist and constructionist psychological theories, we develop interpretable models to predict well-being and identify adaptive and maladaptive self-states in longitudinal social media data. Our approach integrates person-level psychological traits (e.g., resilience, cognitive distortions, implicit motives) with language-inferred situational features derived from the Situational 8 DIAMONDS framework. We compare these theory-grounded features to embeddings from a psychometrically-informed language model that captures temporal and individual-specific patterns. Results show that our principled, theory-driven features provide competitive performance while offering greater interpretability. Qualitative analyses further highlight the psychological coherence of features most predictive of well-being. These findings underscore the value of integrating computational modeling with psychological theory to assess dynamic mental states in contextually sensitive and human-understandable ways.

Charles Welch, Allison Lahnala, Vasudha Varadarajan et al.

We present a low-compute non-generative system for implementing interview-style conversational agents which can be used to facilitate qualitative data collection through controlled interactions and quantitative analysis. Use cases include applications to tracking attitude formation or behavior change, where control or standardization over the conversational flow is desired. We show how our system can be easily adjusted through an online administrative panel to create new interviews, making the tool accessible without coding. Two case studies are presented as example applications, one regarding the Expressive Interviewing system for COVID-19 and the other a semi-structured interview to survey public opinion on emerging neurotechnology. Our code is open-source, allowing others to build off of our work and develop extensions for additional functionality.

Yashwanth YS, Ruichen Wang, Shihua Zeng et al.

As LLM-based agents are increasingly interacting in multi-party settings, they need to properly handle information asymmetry, i.e., knowing when and to whom to disclose information is appropriate. Yet, existing benchmarks fail to measure this ability in realistic multi-party settings. Thus, we introduce SOTOPIA-TOM, a multi-dimensional benchmarking framework to evaluate LLM agents' ability to successfully navigate information asymmetric and privacy sensitive multi-party interactions. We create an interaction environment which enables both public (broadcast) and private (direct message) communication, and craft 160 human-reviewed scenarios across eight industry sectors, each involving 3 to 5 agents with partitioned private knowledge and channel-dependent sharing policies. To measure interaction abilities, we create a multi-dimensional evaluation framework to assess how well agents share useful information, seek missing details, coordinate efficiently, and protect privacy, which we also combine into a composite INFOMGMT metric. Results show that, across 6 LLM backbones and prompting strategies (vanilla, CoT-privacy, and ToM-based interventions), even the largest high-reasoning model (GPT-5) reaches only a 62% INFOMGMT score, which indicates persistent deficiencies in information seeking and privacy-aware decision-making. Additionally, ToM-based interventions more consistently improve the overall coordination-privacy balance (for example, relative to the vanilla baseline, ToM-Coach reduces critical privacy violations on GPT-4o from 9.9% to 2.2% while increasing the composite InfoMgmt score more than 2.5x from 15% to 40%). Overall, SOTOPIA-TOM exposes persistent limitations of current LLM agents in complex, information-asymmetric coordination and provides an extensible testbed for developing more privacy-aware, theory-of-mind capable multi-agent systems.

Allison Lahnala, Vasudha Varadarajan, Lucie Flek et al.

The proliferation of ideological movements into extremist factions via social media has become a global concern. While radicalization has been studied extensively within the context of specific ideologies, our ability to accurately characterize extremism in more generalizable terms remains underdeveloped. In this paper, we propose a novel method for extracting and analyzing extremist discourse across a range of online community forums. By focusing on verbal behavioral signatures of extremist traits, we develop a framework for quantifying extremism at both user and community levels. Our research identifies 11 distinct factors, which we term ``The Extremist Eleven,'' as a generalized psychosocial model of extremism. Applying our method to various online communities, we demonstrate an ability to characterize ideologically diverse communities across the 11 extremist traits. We demonstrate the power of this method by analyzing user histories from members of the incel community. We find that our framework accurately predicts which users join the incel community up to 10 months before their actual entry with an AUC of $>0.6$, steadily increasing to AUC ~0.9 three to four months before the event. Further, we find that upon entry into an extremist forum, the users tend to maintain their level of extremism within the community, while still remaining distinguishable from the general online discourse. Our findings contribute to the study of extremism by introducing a more holistic, cross-ideological approach that transcends traditional, trait-specific models.

Vasudha Varadarajan, Syeda Mahwish, Xiaoran Liu et al.

While NLP models often seek to capture cognitive states via language, the validity of predicted states is determined by comparing them to annotations created without access the cognitive states of the authors. In behavioral sciences, cognitive states are instead measured via experiments. Here, we introduce an experiment-based framework for evaluating language-based cognitive style models against human behavior. We explore the phenomenon of decision making, and its relationship to the linguistic style of an individual talking about a recent decision they made. The participants then follow a classical decision-making experiment that captures their cognitive style, determined by how preferences change during a decision exercise. We find that language features, intended to capture cognitive style, can predict participants' decision style with moderate-to-high accuracy (AUC ~ 0.8), demonstrating that cognitive style can be partly captured and revealed by discourse patterns.

Kieran Henderson, Kian Omoomi, Vasudha Varadarajan et al.

Recent work has explored the use of personal information in the form of persona sentences or self-disclosures to improve modeling of individual characteristics and prediction of annotator labels for subjective tasks. The volume of personal information has historically been restricted and thus little exploration has gone into understanding what kind of information is most informative for predicting annotator labels. In this work, we categorize self-disclosures and use them to build annotator models for predicting judgments of social norms. We perform several ablations and analyses to examine the impact of the type of information on our ability to predict annotation patterns. Contrary to previous work, only a small number of comments related to the original post are needed. Lastly, a more diverse sample of annotator self-disclosures did not lead to the best performance. Sampling from a larger pool of comments without filtering still yields the best performance, suggesting that there is still much to uncover in terms of what information about an annotator is most useful for verdict prediction.

Adithya V Ganesan, Vasudha Varadarajan, Oscar NE Kjell et al.

While NLP typically treats documents as independent and unordered samples, in longitudinal studies, this assumption rarely holds: documents are nested within authors and ordered in time, forming person-indexed, time-ordered $\textit{behavioral sequences}$. Here, we demonstrate the need for and propose a longitudinal modeling and evaluation paradigm that consequently updates four parts of the NLP pipeline: (1) evaluation splits aligned to generalization over people ($\textit{cross-sectional}$) and/or time ($\textit{prospective}$); (2) accuracy metrics separating between-person differences from within-person dynamics; (3) sequence inputs to incorporate history by default; and (4) model internals that support different $\textit{coarseness}$ of latent state over histories (pooled summaries, explicit dynamics, or interaction-based models). We demonstrate the issues ensued by traditional pipeline and our proposed improvements on a dataset of 17k daily diary transcripts paired with PTSD symptom severity from 238 participants, finding that traditional document-level evaluation can yield substantially different and sometimes reversed conclusions compared to our ecologically valid modeling and evaluation. We tie our results to a broader discussion motivating a shift from word-sequence evaluation toward $\textit{behavior-sequence}$ paradigms for NLP.

Vasudha Varadarajan, Hui Xu, Rebecca Astrid Boehme et al.

Recent advances in large language models (LLMs) offer new opportunities for scalable, interactive mental health assessment, but excessive querying by LLMs burdens users and is inefficient for real-world screening across transdiagnostic symptom profiles. We introduce MAQuA, an adaptive question-asking framework for simultaneous, multidimensional mental health screening. Combining multi-outcome modeling on language responses with item response theory (IRT) and factor analysis, MAQuA selects the questions with most informative responses across multiple dimensions at each turn to optimize diagnostic information, improving accuracy and potentially reducing response burden. Empirical results on a novel dataset reveal that MAQuA reduces the number of assessment questions required for score stabilization by 50-87% compared to random ordering (e.g., achieving stable depression scores with 71% fewer questions and eating disorder scores with 85% fewer questions). MAQuA demonstrates robust performance across both internalizing (depression, anxiety) and externalizing (substance use, eating disorder) domains, with early stopping strategies further reducing patient time and burden. These findings position MAQuA as a powerful and efficient tool for scalable, nuanced, and interactive mental health screening, advancing the integration of LLM-based agents into real-world clinical workflows.

Adithya V Ganesan, Vasudha Varadarajan, Yash Kumar Lal et al.

Use of large language models such as ChatGPT (GPT-4/GPT-5) for mental health support has grown rapidly, emerging as a promising route to assess and help people with mood disorders like depression. However, we have a limited understanding of these language models' schema of mental disorders, that is, how they internally associate and interpret symptoms of such disorders. In this work, we leveraged contemporary measurement theory to decode how GPT-4 and GPT-5 interrelate depressive symptoms, providing an explanation of how LLMs apply what they learn and informing clinical applications. We found that GPT-4 (a) had strong convergent validity with standard instruments and expert judgments $(r = 0.70 - 0.81)$, and (b) behaviorally linked depression symptoms with each other (symptom inter-correlates $r = 0.23 - 0.78$) in accordance with established literature on depression; however, it (c) underemphasized the relationship between $\textit{suicidality}$ and other symptoms while overemphasizing $\textit{psychomotor symptoms}$; and (d) suggested novel hypotheses of symptom mechanisms, for instance, indicating that $\textit{sleep}$ and $\textit{fatigue}$ are broadly influenced by other depressive symptoms, while $\textit{worthlessness/guilt}$ is only tied to $\textit{depressed mood}$. GPT-5 showed a slightly lower convergence with self-report, a difference our machine-behavior analysis makes interpretable through shifts in symptom-symptom relationships. These insights provide an empirical foundation for understanding language models' mental health assessments and demonstrate a generalizable approach for explainability in other models and disorders. Our findings can guide key stakeholders to make informed decisions for effectively situating these technologies in the care system.

Vasudha Varadarajan, Swanie Juhng, Syeda Mahwish et al.

While transformer-based systems have enabled greater accuracies with fewer training examples, data acquisition obstacles still persist for rare-class tasks -- when the class label is very infrequent (e.g. < 5% of samples). Active learning has in general been proposed to alleviate such challenges, but choice of selection strategy, the criteria by which rare-class examples are chosen, has not been systematically evaluated. Further, transformers enable iterative transfer-learning approaches. We propose and investigate transfer- and active learning solutions to the rare class problem of dissonance detection through utilizing models trained on closely related tasks and the evaluation of acquisition strategies, including a proposed probability-of-rare-class (PRC) approach. We perform these experiments for a specific rare class problem: collecting language samples of cognitive dissonance from social media. We find that PRC is a simple and effective strategy to guide annotations and ultimately improve model accuracy while transfer-learning in a specific order can improve the cold-start performance of the learner but does not benefit iterations of active learning.

Youngseo Son, Vasudha Varadarajan, H Andrew Schwartz

Discourse relations are typically modeled as a discrete class that characterizes the relation between segments of text (e.g. causal explanations, expansions). However, such predefined discrete classes limits the universe of potential relationships and their nuanced differences. Analogous to contextual word embeddings, we propose representing discourse relations as points in high dimensional continuous space. However, unlike words, discourse relations often have no surface form (relations are between two segments, often with no word or phrase in that gap) which presents a challenge for existing embedding techniques. We present a novel method for automatically creating discourse relation embeddings (DiscRE), addressing the embedding challenge through a weakly supervised, multitask approach to learn diverse and nuanced relations between discourse segments in social media. Results show DiscRE can: (1) obtain the best performance on Twitter discourse relation classification task (macro F1=0.76) (2) improve the state of the art in social media causality prediction (from F1=.79 to .81), (3) perform beyond modern sentence and contextual word embeddings at traditional discourse relation classification, and (4) capture novel nuanced relations (e.g. relations semantically at the intersection of causal explanations and counterfactuals).