We introduce C-QUERI, a dataset of question-answer exchanges from U.S. congressional committee hearings spanning the 108th–117th Congresses, with the pipeline used to construct it. C-QUERI transforms congressional hearing records into structured question-answer exchanges, enabling researchers to study political questioning and witness responses systematically and at scale. Congressional hearings provide a particularly valuable setting for such research: interactions are structured by formal rules, members of different parties are guaranteed opportunities to question witnesses, and witnesses are expected to respond. As an initial application demonstrating the resource’s utility, we examine whether questioner party affiliation can be predicted from question text alone. Question text contains a modest aggregate partisan signal, with substantially greater distinguishability in some committees, hearing types, and configurations of partisan control than others. C-QUERI provides new infrastructure for studying congressional politics and a general framework for analyzing question-answer interactions in other interview-like settings.

Fair allocation of scarce, indivisible resources is an important challenge in many societal problems. While there are several formal theories of fairness, no single definition can always be satisfied. As large language models (LLMs) are increasingly used to support decisions and act as agents, they raise new concerns about distributional justice: their judgments are not directly tied to any specific fairness framework and may violate key normative principles. In this work, we introduce a general method for evaluating fairness reasoning in LLMs. We study first-person fairness judgments across a broad set of models and compare them directly with human responses on matched scenarios and elicitation conditions. We find that LLMs tend to prefer stricter fairness constraints than humans, show more self-interested behavior, are sensitive to how information is framed, and are difficult to align with human judgments using fine-tuning with current datasets.