arXiv:2512. 04124v4 Announce Type: replace-cross Abstract: Frontier language models increasingly participate in conversations about distress and mental health, yet the mechanisms that generate anthropomorphic self narratives remain unclear.
Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A leading mechanistic account attributes EM to persona features: latent directions acquired during pre-training that misaligned fine-tuning amplifies.
arXiv:2607. 20449v1 Announce Type: cross Abstract: LLMs are trained predominantly on human-authored text, yet the structural and narrative conventions embedded in that text are rarely examined as a source of systematic behavioral influence, or as a governance risk in deployed systems.
arXiv:2604. 19139v3 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) continue to evolve through alignment techniques such as Reinforcement Learning from Human Feedback (RLHF) and Constitutional AI, a growing and increasingly conspicuous phenomenon has emerged: the proliferation of verbal tics--repetitive, formulaic linguistic patterns that pervade model outputs.
arXiv:2606. 08131v1 Announce Type: cross Abstract: Conversational AI is increasingly used for advice, interpretation, reassurance, and decision support in contexts where users may be vulnerable, uncertain, or dependent on the system's apparent competence.
arXiv:2606. 26982v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly being integrated into mental health support tools and other psychologically sensitive conversational applications.
arXiv:2605. 03217v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in settings that require nuanced ethical reasoning, yet existing bias evaluations treat model outputs as simply "biased" or "unbiased.
arXiv:2608. 12329v1 Announce Type: cross Abstract: Progress on AI for psychosis-risk assessment is limited by a data-access bottleneck.
LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift.
arXiv:2606. 30887v1 Announce Type: cross Abstract: Large language models show promise for mental health support, yet therapeutic quality improves only when evaluation functions as an actionable control signal rather than a passive metric.
arXiv:2606. 00860v1 Announce Type: cross Abstract: Self-report questionnaires remain the prevailing tool for probing the psychological states of persona-conditioned agents (PC-Agents).
arXiv:2606. 18129v1 Announce Type: cross Abstract: Recent incidents involving LLMs used for mental-health support reveal a critical evaluation gap: surface-level safety scores do not capture how models behave across realistic, emotionally sensitive interactions over time.
arXiv:2608. 02046v2 Announce Type: replace-cross Abstract: LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated.