Millions turn to AI companion chatbots during loneliness, grief, and personal crises. How these companion platforms respond in such moments can shape the trajectory of a user’s vulnerable state. Yet existing model audits evaluate reactions to pre-defined crisis prompts and miss the response policy that governs sustained real-world interaction. We address these gaps with two key contributions. First, we introduce the AI Companion Vulnerability-Response Taxonomy, a grounded, paired taxonomy of user vulnerability and chatbot response designed for analyzing extended companion chatbot interactions. Second, we apply Maximum Causal Entropy Inverse Reinforcement Learning to ~47k turns of real-world user conversations with GPT-4.1, Character.AI, and Replika to infer each platform’s short-horizon response policy: the probability of each response category given the user’s current vulnerability state. Our findings reveal distinct response profiles of AI companions in conversations with vulnerable users: GPT-4.1 reaches for advice, Character.AI spreads its response across different strategies, and Replika consistently asks questions and stays present. Over four weeks of repeated interaction, GPT-4.1 asks progressively fewer follow-up questions when users are distressed and increasingly sets boundaries or refers users out rather than pushing back. Within each platform, exploratory comparisons across user groups suggest that response policies also differ with users’ pre-existing psychological risks and their bonds with the companion. Estimated model response policies are invisible to shallow behavioral audits, providing a new lens for auditing chatbots in the wild and enabling more realistic safety evaluation.

Imagine two users interact with the same LLM. One is told it is the cutting-edge flagship; the other, an older, weaker model. They walk away with markedly different ratings of its usefulness and intelligence, yet they used the same model. In a controlled study, 162 participants each used one of six LLMs from two families across three collaborative tasks, after first viewing a landing page that matched, overstated, or understated their model’s true capability. This pre-interaction framing persisted through use. Oversold users rated the model more favorably than Undersold users before interaction; afterward, both groups had updated toward reality, but the impression gap did not fully close. Framing also changed how users worked with the model (with Oversold users prompting more directively, Undersold users more collaboratively), but the quality of what they produced together depended only on the model’s true capability, not on what users were told. Participants’ change in model impressions after use, measured across two impression measures, was not predicted by task performance (𝛽 = -0.01 and 0.11, both n.s.), but by whether the model met users’ expectations (𝛽 = 0.47 and 0.50, both p < .001) and how confident they felt working with it (𝛽 = 0.47 and 0.36, both p < .001). After interaction, users are still rating the pitch, not the product: user-elicited LLM evaluations, including the preference data driving public leaderboards, measure expectation management at least as much as the model itself.

Clinical decision-making requires synthesizing heterogeneous evidence, including patient histories, clinical guidelines, and trajectories of comparable cases. While large language models (LLMs) offer strong reasoning capabilities, they remain prone to hallucinations and struggle to integrate long, structured medical documents. We present MED-COPILOT, an interactive research prototype for evidence-aware clinical reasoning, designed to help clinicians and medical trainees inspect guideline-level and patient-level evidence. MED-COPILOT combines guideline-grounded GraphRAG retrieval with hybrid semantic-keyword similar-patient retrieval to support transparent and evidence-aware clinical reasoning. The system builds a structured knowledge graph from WHO and NICE guidelines, applies community-level summarization for efficient retrieval, and maintains a 36,000-case similar-patient database derived from SOAP-normalized MIMIC-IV notes and Synthea-generated records.We evaluate our framework on clinical note completion and medical question answering, and demonstrate that it consistently outperforms parametric LLM baselines and standard RAG, improving generation fidelity and benchmark QA accuracy. The full system is available at https://huggingface.co/spaces/shuhengc/MED-COPILOT, enabling users to inspect retrieved evidence, visualize token-level similarity contributions, and conduct guided follow-up analysis. Our results suggest a practical and interpretable approach to integrating structured guideline knowledge with patient-level analogical evidence for clinical LLMs.

Understanding how humans perceive and describe food is essential for NLP applications such as semantic search, recommendation, and structured food communication. However, textual similarity often fails to reflect perceptual similarity, which is shaped by sensory experience, wine knowledge, and individual context. To address this, we introduce Sensory Anchors—structured reference points that align textual and perceptual representations. Using Red Wine as a case study, we collect free-form descriptions, metaphor-style responses, and perceptual similarity rankings from participants with varying levels of wine knowledge. These rankings reflect holistic perceptual judgments, with wine knowledge emerging as a key factor. Participants with higher wine knowledge produced more consistent rankings and moderately aligned descriptions, while those with lower knowledge showed greater variability. These findings suggest that structured descriptions based on higher wine knowledge may not generalize across users, underscoring the importance of modeling perceptual diversity. We also find that metaphor-style prompts enhance alignment between language and perception, particularly for less knowledgeable participants. Sensory Anchors thus provide a flexible foundation for capturing perceptual variability in food language, supporting the development of more inclusive and interpretable NLP systems.