TPvG: A Moral Decision Framework for Large Language Models from One-Shot to Sequential Feedback

Researchers introduced TPvG (Text-based Pain-versus-Gain), a moral decision framework for large language models that embeds consequence feedback into moral dilemmas, and found that LLM moral decisions were strongly affected by decision format (one-shot versus sequential) and that explicit receiver feedback produced heterogeneous effects across models, diverging from human reference patterns. The study, posted on arXiv (2608.28610v1), highlights the need to evaluate LLM moral behavior in high-stakes interactive settings.

arXiv:2608.28610v1 Announce Type: new Abstract: Existing LLM moral evaluations typically present models with isolated moral vignettes and elicit a single-shot decision, neglecting a factor known to profoundly influence human moral behavior: consequence feedback. We introduce TPvG Text-based Pain-versus-Gain , adapted from a human moral paradigm, which embeds consequence feedback into an everyday moral dilemma of not harming others versus maximising self-gain. TPvG comprises five moral decision tasks, progressing from minimal-context one-shot choices to sequential decisions with explicit consequence feedback. Our results show that LLM moral decisions were strongly affected by decision format one-shot versus sequential , and explicit receiver feedback produced heterogeneous effects across models. Furthermore, LLM responses to explicit receiver feedback diverged from the human reference pattern, suggesting potentially different decision processes. These findings highlight the need to evaluate whether LLM moral behavior remains stable in high-stakes interactive settings.