← Back to all news
arXiv Computation and Language By Yuxiang Gao, Tianjie Ju, Zheng Wu, Zongru Wu, Hao Fei

ACTOR: Action-Centered Temporal Organization for Literary Role-Playing Agents

Read the original on arXiv Computation and Language →

The paper introduces ACTOR, a framework that automatically constructs action trajectories for literary role‑playing agents by extracting dialogue and non‑dialogue behaviors from novels and representing them with a Thought‑Action‑Observation structure. It builds a dataset from 143 literary works, generates counterfactual behaviors, and trains models using Regularized Preference Optimization to improve character‑consistent action prediction. Evaluation on 17,560 instances shows that the ACTOR‑trained model significantly outperforms baselines in next‑behavior selection, demonstrating the value of temporally organized supervision for literary role‑playing.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

Related stories

arXiv AI
Sep 25

Adversarial Closed-Loop Curriculum for Evolving Role-Playing Agents

The paper introduces AdvRole, an adversarial closed‑loop curriculum for training role‑playing agents with large language models. It alternates between an Actor that learns to role‑play and a Rewriter that edits character profiles and dialogue contexts into hard scenarios, using a performance‑gap reward to target the Actor’s weaknesses. Experiments on English, Chinese, and a new multilingual benchmark demonstrate that AdvRole consistently outperforms baseline methods.

By Zheng Zhang, Liu Liu, Qi Chai, Deheng Ye, Peilin Zhao, Mao Zheng, Hao Wang
llmsagentsreinforcement-learningbenchmarkssafety
More like this →
arXiv Computation and Language
2d ago

Steering Readerly Dynamics: Trajectory-Guided Story Generation with Large Language Models

The paper introduces a framework that learns readerly dynamics—how readers respond over time—from high‑quality human stories and uses these dynamics to steer large language model (LLM) story generation. Instead of modeling full story content, a small LLM predicts low‑dimensional trajectories of future reader responses, which a larger, instruction‑following writer LLM then follows to produce story segments. Experiments show the approach improves trajectory‑based fidelity to human stories by 12%–35% while largely maintaining coherence.

By Danrui Li, Yiwei Zhao, Mubbasir Kapadia, Lingxiao Li
llms
More like this →