Implicitly conveying a user's professional role in multi-turn LLM conversations shifts moral wrongness ratings across ten common-morality rules in two non-reasoning models.
Inertia in Moral and Value Judgments of Large Language Models
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
Large Language Models (LLMs) behave non-deterministically, and prompting has become a common method for steering their outputs. A popular strategy is to assign a persona to the model to produce more varied, context-sensitive responses, similar to how responses vary across human individuals. Against the expectation that persona prompting yields a wide range of opinions, our experiments show that LLMs keep consistent value orientations. We observe a persistent inertia in their responses, where certain moral and value dimensions (especially harm avoidance and fairness) stay skewed in one direction across persona settings. To study this, we use role-play at scale, which pairs randomized persona prompts with a macro-level analysis of model outputs. Our results point to strong internal biases and value preferences in LLMs, which we call value orientation and inertia. These models warrant scrutiny and adjustment before use in applications where balanced outputs matter.
verdicts
CONDITIONAL 3representative citing papers
XtraGPT is a suite of 1.5B-14B parameter open-source LLMs fine-tuned on 140,000 revision pairs from 7,000 top-tier papers to support controllable, context-aware academic paper editing.
Standard psychometric questionnaires like the Big Five and PVQ produce different and more consistent results than ecologically valid questions drawn from real user conversations, suggesting the former may mischaracterize LLM behavior.
citing papers explorer
-
User identity conditions moral wrongness ratings in non-reasoning large language models
Implicitly conveying a user's professional role in multi-turn LLM conversations shifts moral wrongness ratings across ten common-morality rules in two non-reasoning models.
-
XtraGPT: Context-Aware and Controllable Academic Paper Revision via Human-AI Collaboration
XtraGPT is a suite of 1.5B-14B parameter open-source LLMs fine-tuned on 140,000 revision pairs from 7,000 top-tier papers to support controllable, context-aware academic paper editing.
-
Human Psychometric Questionnaires Mischaracterize LLM Behavior
Standard psychometric questionnaires like the Big Five and PVQ produce different and more consistent results than ecologically valid questions drawn from real user conversations, suggesting the former may mischaracterize LLM behavior.