ANIMASK: What the Model Contributes to Role Play in Simulated Story Worlds

• URL: http://arxiv.org/abs/2609.16667v1
• Date: Tue, 15 Sep 2026 05:43:35 GMT
• Title: ANIMASK: What the Model Contributes to Role Play in Simulated Story Worlds
• Abstract summary: ANIMASK is a framework that freezes books and scripts and replays each story from its freeze point.<n>Across 40 stories, 6 actor models, and 3,846 decision points, the replays converge away from their canons in one shared direction.
• Score: 7.891251085330825
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: When a language model plays a character, the observed behavior reflects both the assigned persona and the default dispositions of the actor model itself. Existing evaluations test persona fidelity or model defaults in isolation, but neither says, at a specific choice with consequences, what the persona changed and what the model's default kept. We introduce ANIMASK, a simulation framework that freezes books and scripts into story worlds whose characters act on their own motivations and replays each story from its freeze point. We hold out the author's continuation as a human reference, verify through in-story interviews that each persona remains present, and at every decision point compare the character's action with what the model produces when the persona is removed. Across 40 stories, 6 actor models, and 3,846 decision points, the replays converge away from their canons in one shared direction, toward flatter, cooler stories that leave their tensions open. The personas stay present and obeyed throughout. On three choices in four the model's default already falls inside what the persona accepts, and where the two diverge the model is the cautious one, holding where the persona would press. The persona guarantees who the character is, and the model sets how far the character will go.

Related papers

• NARRA-Gym for Evaluating Interactive Narrative Agents [69.49891044929372]
  We introduce NARRA-Gym, an evaluation environment that turns a sparse emotional seed into a complete interactive story episode.<n>We evaluate nine frontier LLMs using a controlled LLM-as-judge sweep over eight benchmark personas and a human evaluation in which participants rate customized model outputs.
  arXiv  Detail & Related papers  (2026-05-08T21:36:23Z)
• StoryAlign: Evaluating and Training Reward Models for Story Generation [52.44891270363296]
  We introduce StoryRMB, the first benchmark for assessing reward models on story preferences.<n>We develop StoryReward, an advanced reward model for story preference trained on this dataset.<n>StoryReward achieves state-of-the-art (SoTA) performance on StoryRMB, outperforming much larger models.
  arXiv  Detail & Related papers  (2026-05-06T12:28:17Z)
• Breaking the Assistant Mold: Modeling Behavioral Variation in LLM Based Procedural Character Generation [62.54606886226136]
  Procedural content generation has enabled vast virtual worlds through levels, maps, and quests, but large-scale character generation remains underexplored.<n>We identify two alignment-induced biases in existing methods.<n>We introduce PersonaWeaver, a framework that disentangles world-building from behavioral-building.
  arXiv  Detail & Related papers  (2026-01-06T20:18:01Z)
• Assessing Social Alignment: Do Personality-Prompted Large Language Models Behave Like Humans? [9.771036970279765]
  State-of-the-art approaches exploit a vast variety of training data, and prompt the model to adopt a particular personality.<n>We use classic psychological experiments, the Milgram experiment and the Ultimatum Game, as social interaction testbeds.<n>Our experiments reveal failure modes of the prompt-based modulation of the models' behavior that are shared across all models tested and persist under prompt perturbations.
  arXiv  Detail & Related papers  (2024-12-21T20:58:19Z)
• Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systems [103.416202777731]
  We study "persona biases", which we define to be the sensitivity of dialogue models' harmful behaviors contingent upon the personas they adopt. We categorize persona biases into biases in harmful expression and harmful agreement, and establish a comprehensive evaluation framework to measure persona biases in five aspects: Offensiveness, Toxic Continuation, Regard, Stereotype Agreement, and Toxic Agreement.
  arXiv  Detail & Related papers  (2023-10-08T21:03:18Z)
• Skeleton-free Pose Transfer for Stylized 3D Characters [53.33996932633865]
  We present the first method that automatically transfers poses between stylized 3D characters without skeletal rigging. We propose a novel pose transfer network that predicts the character skinning weights and deformation transformations jointly to articulate the target character to match the desired pose. Our method is trained in a semi-supervised manner absorbing all existing character data with paired/unpaired poses and stylized shapes.
  arXiv  Detail & Related papers  (2022-07-28T20:05:57Z)
• Evaluating and Inducing Personality in Pre-trained Language Models [78.19379997967191]
  We draw inspiration from psychometric studies by leveraging human personality theory as a tool for studying machine behaviors. To answer these questions, we introduce the Machine Personality Inventory (MPI) tool for studying machine behaviors. MPI follows standardized personality tests, built upon the Big Five Personality Factors (Big Five) theory and personality assessment inventories. We devise a Personality Prompting (P2) method to induce LLMs with specific personalities in a controllable way.
  arXiv  Detail & Related papers  (2022-05-20T07:32:57Z)
• Persona-Guided Planning for Controlling the Protagonist's Persona in Story Generation [71.24817035071176]
  We propose a planning-based generation model named CONPER to explicitly model the relationship between personas and events. Both automatic and manual evaluation results demonstrate that CONPER outperforms state-of-the-art baselines for generating more coherent and persona-controllable stories.
  arXiv  Detail & Related papers  (2022-04-22T13:45:02Z)
• Immersive Text Game and Personality Classification [1.9171404264679484]
  Immersive Text Game allows the player to choose a story and a character, and interact with other characters in the story in an immersive manner. The game is based on several latest models, including text generation language model, information extraction model, commonsense reasoning model, and psychology evaluation model.
  arXiv  Detail & Related papers  (2022-03-20T18:37:03Z)
• Triangular Character Animation Sampling with Motion, Emotion, and Relation [78.80083186208712]
  We present a novel framework to sample and synthesize animations by associating the characters' body motions, facial expressions, and social relations. Our method can provide animators with an automatic way to generate 3D character animations, help synthesize interactions between Non-Player Characters (NPCs) and enhance machine emotion intelligence in virtual reality (VR)
  arXiv  Detail & Related papers  (2022-03-09T18:19:03Z)
• Telling Stories through Multi-User Dialogue by Modeling Character Relations [14.117921448623342]
  This paper explores character-driven story continuation, in which the story emerges through characters' first- and second-person narration as well as dialogue. We hypothesize that a multi-task model that trains on character dialogue plus character relationship information improves transformer-based story continuation. A series of ablations lend evidence to our hypothesis, showing that our multi-task model using character relationships improves story continuation accuracy over strong baselines.
  arXiv  Detail & Related papers  (2021-05-31T15:39:41Z)
• Controllable Multi-Character Psychology-Oriented Story Generation [28.054245616281023]
  We present a novel model-based attention mechanism that we call SoCP (Storytelling of multi-Character Psychology) We show that the proposed model can generate stories considering the changes in the psychological state of different characters. Experiments show that with SoCP, the generated stories follow the psychological state for each character according to both automatic and human evaluations.
  arXiv  Detail & Related papers  (2020-10-11T12:05:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.