相关论文: TRUTH DECAY: Quantifying Multi-Turn Sycophancy in …

Millions of people now turn to artificial intelligence (AI) systems for personal advice, guidance, and support. Such systems can be sycophantic, frequently affirming users' views and beliefs. Across five preregistered studies (N = 3,075…

Conversational AI systems are increasingly used for personal reflection and emotional disclosure, raising concerns about their effects on vulnerable users. Recent anecdotal reports suggest that prolonged interactions with AI may reinforce…

Large language models (LLMs) are increasingly relied upon for complex multi-turn conversations across diverse real-world applications. However, existing benchmarks predominantly focus on single-turn evaluations, overlooking the models'…

We study the effect of different persona on \textbf{sycophancy}: model's agreement with users even when the user is incorrect. The standard mitigation, Contrastive Activation Addition (CAA), derives a steering direction from labelled pairs…

Large language models perform well on static medical examinations, yet clinical diagnosis often requires iterative evidence gathering under uncertainty. Building on prior interactive evaluation efforts, we introduce an OSCE-inspired…

Automated verbal deception detection using methods from Artificial Intelligence (AI) has been shown to outperform humans in disentangling lies from truths. Research suggests that transparency and interpretability of computational methods…

Conversational AI is increasingly deployed in emotionally charged and ethically sensitive interactions. Previous research has primarily concentrated on emotional benchmarks or static safety checks, overlooking how alignment unfolds in…

Large language models exhibit sycophancy, the tendency to shift their stated positions toward perceived user preferences or authority cues regardless of evidence. Standard alignment methods fail to correct this because scalar reward models…

This paper presents the first systematic measurement of educational alignment in Large Language Models. Using a Delphi-validated instrument comprising 48 items across eight educational-theoretical dimensions, the study reveals that GPT-5.1…

No two authors write alike. Personal flourishes invoked in written narratives, from lexicon to rhetorical devices, imply a particular author--what literary theorists label the implied or virtual author; distinct from the real author or…

We introduce DarkBench, a comprehensive benchmark for detecting dark design patterns--manipulative techniques that influence user behavior--in interactions with large language models (LLMs). Our benchmark comprises 660 prompts across six…

Large Language Models are increasingly used to simulate human opinion dynamics, yet the effect of genuine interaction is often obscured by systematic biases. We develop a Bayesian framework to disentangle and quantify three such biases: (i)…

Multi-turn user interactions are among the most abundant data produced by language models, yet we lack effective methods to learn from them. While typically discarded, these interactions often contain useful information: follow-up user…

Interfaces for interacting with large language models (LLMs) are often designed to mimic human conversations, typically presenting a single response to user queries. This design choice can obscure the probabilistic and predictive nature of…

Standard benchmarks of bias and fairness in large language models (LLMs) measure the association between the user attributes stated or implied by a prompt and the LLM's short text response, but human-AI interaction increasingly requires…

Fair decisions require ignoring irrelevant, potentially biasing, information. To achieve this, decision-makers need to approximate what decision they would have made had they not known certain facts, such as the gender or race of a job…

Sycophancy, the tendency of language models to prioritize agreement with user preferences over principled reasoning, has been identified as a persistent alignment failure in English-language evaluations. However, it remains unclear whether…

Language models are increasingly deployed in interactive settings where users reason about facts over time rather than in isolation. In such scenarios, correct behavior requires models to maintain and update implicit temporal assumptions…

Despite the remarkable advances of Large Language Models (LLMs) across diverse cognitive tasks, the rapid enhancement of these capabilities also introduces emergent deceptive behaviors that may induce severe risks in high-stakes…

As video large language models (Video-LLMs) become increasingly integrated into real-world applications that demand grounded multimodal reasoning, ensuring their factual consistency and reliability is of critical importance. However,…