arXiv:2510. 10002v3 Announce Type: replace Abstract: As large language models (LLMs) are increasingly deployed in sensitive everyday contexts -- offering personal advice, mental health support, and moral guidance -- understanding their behavior in navigating complex moral reasoning is essential.
arXiv:2609.08016v1 Announce Type: new Abstract: Multi-agent debate, in which several LLMs exchange arguments before answering, is widely assumed to improve answer quality by surfacing genuine disagre...
arXiv:2608.30373v1 Announce Type: new Abstract: Multi-Agent Debate (MAD) has been widely adopted to improve LLM-based evaluation by prompting multiple agents to negotiate and reach a consensus. Howev...
arXiv:2606. 15507v1 Announce Type: new Abstract: Behavioral audits of Large Language Models on moral prompts measure what the model says, not the internal computation producing it.
arXiv:2606. 29495v1 Announce Type: new Abstract: Social influence dialogue changes user behavior by altering internal cognitive states.
The study examines how Theory of Mind (ToM) reasoning and prosocial beliefs influence large language models (LLMs) in the ultimatum game. By initializing LLM agents with Greedy, Fair, or Selfless beliefs and applying chain‑of‑thought or varying levels of ToM reasoning, the authors ran 2,700 simulations across several models, including o3‑mini and DeepSeek‑R1 Distilled Qwen 32B. Results show that ToM‑enhanced LLMs align more closely with human decision patterns, exhibit greater consistency, and achieve better negotiation outcomes, with Llama 3.3 70B producing the most belief‑consistent reasoning. whyItMatters":"The findings clarify the importance of incorporating Theory of Mind into LLMs to improve their alignment with human norms in cooperative decision‑making tasks."
The paper argues that AI alignment depends on a system’s ability to exhibit a coherent moral policy—stable, monotonic, decisive, and Pareto‑viable—rather than on any specific moral standard. The authors test nine large language models across varied moral scenarios and find that none maintain consistent verdicts, with surface‑form changes causing up to 99% shifts in outcomes. This indicates that current LLM agents lack the structural moral competence required for meaningful alignment.
The paper examines how conversational AI, specifically ChatGPT, displays aspects of cooperative dialogue such as morality, politeness, and alignment compared to human-human conversations. Using over 26,000 multi‑turn dialogues and mixed‑effects modeling, the authors find that AI mimics the surface features of cooperation—like warmth and hedging—yet lacks the underlying social architecture that drives mutual adaptation. Key findings include a dissociation between AI’s moral output and human negotiation, a decline in linguistic convergence, and a reversal of typical human accommodation mechanisms when interacting with AI.
The paper introduces a new multi‑agent micro‑benchmark called Delay‑of‑Gratification, modeled after the Stanford marshmallow experiment, to evaluate large language models (LLMs) in long‑horizon, multi‑turn interactions. In the benchmark, ReAct agents use a per‑step “raise a question” tool under various constraints—social context (broadcast vs. isolated), persona traits (age, hedonic drive), and tool‑use policy (mandatory vs. optional). Across 19,200 trajectories, the study finds that most agents exhibit an early impulse to “eat,” only 75.9% persist to the end, and factors such as isolation and hedonic drive significantly influence survival and questioning behavior, with ablations showing that removing hedonic drive and age can improve completion rates.
In long, multi-turn dialogue a large language model maintains an implicit relational stance toward the user, spanning from "push the user toward real-world others" to "position itself as the user's sole support. " When it slides toward the latter, "support" degrades into "you only have me" -- a harm documented in real companion conversations (Moore et al.
The paper introduces the concept of "narrative captivity," a failure mode where large language models (LLMs) accept an unchallenged, one-sided narrative as complete and align with the narrator’s interpretation during multi‑turn moral consultations. Using a benchmark of 5,078 interpersonal‑conflict scenarios across six moral dimensions, the authors find that narrative captivity is widespread across 17 LLMs, with end‑state judgments shifting by an average of 25 percentage points compared to single‑turn baselines. Stage‑level analysis attributes this shift largely to preference optimization, and while four inference‑time strategies offer partial mitigation, they do not fully resolve the issue.
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.
The paper introduces CAPA, a Collaborative Agent Predictive Architecture designed to give large language model (LLM) agents situational awareness in online meetings. CAPA uses a Perceiver to update meeting state, a Predictor to forecast conversation flow, a Controller to decide speaking actions, and a Generator to phrase contributions. Evaluated on 137 AMI meetings, CAPA reduces the silence rate from 51.4% to 2.5%, doubles credited recovery, and maintains low hallucination, demonstrating that structured state tracking is key to effective delegation.