As large language models (LLMs) are deployed globally, it is crucial that their responses are calibrated across languages to accurately convey uncertainty and limitations. Prior work shows that LLMs are linguistically overconfident in English, leading users to overrely on confident generations. However, the usage and interpretation of epistemic markers (e.g., 'I think it's') differs sharply across languages. Here, we study the risks of multilingual linguistic (mis)calibration, overconfidence, and overreliance across five languages to evaluate LLM safety in a global context. Our work finds that overreliance risks are high across languages. We first analyze the distribution of LLM-generated epistemic markers and observe that LLMs are overconfident across languages, frequently generating strengtheners even as part of incorrect responses. Model generations are, however, sensitive to documented cross-linguistic variation in usage: for example, models generate the most markers of uncertainty in Japanese and the most markers of certainty in German and Mandarin. Next, we measure human reliance rates across languages, finding that reliance behaviors differ cross-linguistically: for example, participants are significantly more likely to discount expressions of uncertainty in Japanese than in English (i.e., ignore their 'hedging' function and rely on generations that contain them). Taken together, these results indicate a high risk of reliance on overconfident model generations across languages. Our findings highlight the challenges of multilingual linguistic calibration and stress the importance of culturally and linguistically contextualized model safety evaluations.

We introduce a new task called Defeasible Visual Entailment (DVE), where the goal is to allow the modification of the entailment relationship between an image premise and a text hypothesis based on an additional update. While this concept is well-established in Natural Language Inference, it remains unexplored in visual entailment. At a high level, DVE enables models to refine their initial interpretations, leading to improved accuracy and reliability in various applications such as detecting misleading information in images, enhancing visual question answering, and refining decision-making processes in autonomous systems. Existing metrics do not adequately capture the change in the entailment relationship brought by updates. To address this, we propose a novel inference-aware evaluator designed to capture changes in entailment strength induced by updates, using pairwise contrastive learning and categorical information learning. Additionally, we introduce a reward-driven update optimization method to further enhance the quality of updates generated by multimodal models. Experimental results demonstrate the effectiveness of our proposed evaluator and optimization method.

In line with the principle of honesty, there has been a growing effort to train large language models (LLMs) to generate outputs containing epistemic markers. However, evaluation in the presence of epistemic markers has been largely overlooked, raising a critical question: Could the use of epistemic markers in LLM-generated outputs lead to unintended negative consequences? To address this, we present EMBER, a benchmark designed to assess the robustness of LLM-judges to epistemic markers in both single and pairwise evaluation settings. Our findings, based on evaluations using EMBER, reveal that all tested LLM-judges, including GPT-4o, show a notable lack of robustness in the presence of epistemic markers. Specifically, we observe a negative bias toward epistemic markers, with a stronger bias against markers expressing uncertainty. This suggests that LLM-judges are influenced by the presence of these markers and do not focus solely on the correctness of the content.

The reconfiguration of human-LM interactions from simple sentence completions to complex, multi-domain, humanlike engagements necessitates new methodologies to understand how humans choose to rely on LMs. In our work, we contend that reliance is influenced by numerous factors within the interactional context of a generation, a departure from prior work that used verbalized confidence (e.g., "I'm certain the answer is...") as the key determinant of reliance. Here, we introduce Rel-A.I., an in situ, system-level evaluation approach to measure human reliance on LM-generated epistemic markers (e.g., "I think it's..", "Undoubtedly it's..."). Using this methodology, we measure reliance rates in three emergent human-LM interaction settings: long-term interactions, anthropomorphic generations, and variable subject matter. Our findings reveal that reliance is not solely based on verbalized confidence but is significantly affected by other features of the interaction context. Prior interactions, anthropomorphic cues, and subject domain all contribute to reliance variability. An expression such as, "I'm pretty sure it's...", can vary up to 20% in reliance frequency depending on its interactional context. Our work underscores the importance of context in understanding human reliance and offers future designers and researchers with a methodology to conduct such measurements.

As natural language becomes the default interface for human-AI interaction, there is a critical need for LMs to appropriately communicate uncertainties in downstream applications. In this work, we investigate how LMs incorporate confidence about their responses via natural language and how downstream users behave in response to LM-articulated uncertainties. We examine publicly deployed models and find that LMs are unable to express uncertainties when answering questions even when they produce incorrect responses. LMs can be explicitly prompted to express confidences, but tend to be overconfident, resulting in high error rates (on average 47%) among confident responses. We test the risks of LM overconfidence by running human experiments and show that users rely heavily on LM generations, whether or not they are marked by certainty. Lastly, we investigate the preference-annotated datasets used in RLHF alignment and find that humans have a bias against texts with uncertainty. Our work highlights a new set of safety harms facing human-LM interactions and proposes design recommendations and mitigating strategies moving forward.

The increased deployment of LMs for real-world tasks involving knowledge and facts makes it important to understand model epistemology: what LMs think they know, and how their attitudes toward that knowledge are affected by language use in their inputs. Here, we study an aspect of model epistemology: how epistemic markers of certainty, uncertainty, or evidentiality like "I'm sure it's", "I think it's", or "Wikipedia says it's" affect models, and whether they contribute to model failures. We develop a typology of epistemic markers and inject 50 markers into prompts for question answering. We find that LMs are highly sensitive to epistemic markers in prompts, with accuracies varying more than 80%. Surprisingly, we find that expressions of high certainty result in a 7% decrease in accuracy as compared to low certainty expressions; similarly, factive verbs hurt performance, while evidentials benefit performance. Our analysis of a popular pretraining dataset shows that these markers of uncertainty are associated with answers on question-answering websites, while markers of certainty are associated with questions. These associations may suggest that the behavior of LMs is based on mimicking observed language use, rather than truly reflecting epistemic uncertainty.

Context is everything, even in commonsense moral reasoning. Changing contexts can flip the moral judgment of an action; "Lying to a friend" is wrong in general, but may be morally acceptable if it is intended to protect their life. We present ClarifyDelphi, an interactive system that learns to ask clarification questions (e.g., why did you lie to your friend?) in order to elicit additional salient contexts of a social or moral situation. We posit that questions whose potential answers lead to diverging moral judgments are the most informative. Thus, we propose a reinforcement learning framework with a defeasibility reward that aims to maximize the divergence between moral judgments of hypothetical answers to a question. Human evaluation demonstrates that our system generates more relevant, informative and defeasible questions compared to competitive baselines. Our work is ultimately inspired by studies in cognitive science that have investigated the flexibility in moral cognition (i.e., the diverse contexts in which moral rules can be bent), and we hope that research in this direction can assist both cognitive and computational investigations of moral judgments.