Operation Veja: Fixing Fundamental Concepts Missing from Modern Roleplaying Training Paradigms
This addresses the issue of creating more authentic roleplaying agents for narrative applications, but it appears incremental as it focuses on data curation rather than a new model architecture.
The paper tackles the problem of modern roleplaying models failing to capture believable characters by identifying missing core concepts (Values, Experiences, Judgments, Abilities) and proposes the VEJA framework for data curation, with a pilot study showing a significant quality gap compared to a state-of-the-art synthetic baseline using LLM-as-judge evaluation.
Modern roleplaying models are increasingly sophisticated, yet they consistently struggle to capture the essence of believable, engaging characters. We argue this failure stems from training paradigms that overlook the dynamic interplay of a character's internal world. Current approaches, including Retrieval-Augmented Generation (RAG), fact-based priming, literature-based learning, and synthetic data generation, exhibit recurring limitations in modeling the deliberative, value-conflicted reasoning that defines human interaction. In this paper, we identify four core concepts essential for character authenticity: Values, Experiences, Judgments, and Abilities (VEJA). We propose the VEJA framework as a new paradigm for data curation that addresses these systemic limitations. To illustrate the qualitative ceiling enabled by our framework, we present a pilot study comparing a manually curated, VEJA-grounded dataset against a state-of-the-art synthetic baseline. Using an LLM-as-judge evaluation, our findings demonstrate a significant quality gap, suggesting that a shift toward conceptually grounded data curation, as embodied by VEJA, is necessary for creating roleplaying agents with genuine depth and narrative continuity. The full dataset is available at https://github.com/HyouinKyoumaIRL/Operation-Veja