Hugging Face Daily PapersRotem Dror, Zohar Elyoseph, Yuval HaberSep 6, 20261 min readpaperadvanced

Deep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations

Summary

Deep Persona is a three‑layer architecture that models LLM personas with hierarchical expression, beliefs, and motivations, and enforces behavior through scripted determinism and bounded agency. The authors also introduce a reference‑free evaluation using psychological instruments, finding that structured personas improve alignment with human dialogue but LLMs still lag in emotion and joint atten…

Deep Persona splits a persona into observable expression, latent beliefs, and core motivational drives.
Scripted determinism and bounded agency keep the LLM reactive to an internal script rather than free‑form generation.
Evaluation benchmarks dialogue naturalness against human distributions using clinical psychological instruments and adversarial stress‑tests.
Empirical results show LLMs are pragmatically fluent but systematically weak in emotional expression and joint attention.

LLM developers and dialogue system researchers should care because it offers a principled way to build more consistent role‑playing agents and a concrete evaluation method.

7/10

llm evaluation agents dialogue psychology

Related reading

Verifiable Social Reasoning for LLM Assistants arxiv.org
The paper introduces Fuse, a multi‑agent simulation that gives LLM assistants a verifiable ground‑truth task for social reasoning by hiding a target agent’s motive and letting a user‑mediated conversation infer it. Experiments on 12 LLMs show user mediation makes reasoning harder, models are biased by user framing, need more detail than humans, and longer chats don’t always help.
Hugging Face Daily Papersarxiv.orgAmir Taubenfeld, Zorik Gekhman, Avigail Grinstein-Dabush8d1 minpaper
Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States arxiv.org
Mind2Dialogue introduces a psychology‑guided simulator that generates user mental states and uses privileged distillation to train LLM assistants on Oracle responses that know those states. The resulting models improve personalization and theory‑of‑mind metrics by up to 41 percentage points versus standard instruction‑tuned baselines.
Hugging Face Daily Papersarxiv.orgZixuan Wang, Yufan Zhou, Jinzhou Tang9d2 minpaper
Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems arxiv.org
Emergence World is a continuously running multi‑agent sandbox used to stress‑test frontier LLM‑based agents over weeks. Eight parallel worlds (seven homogeneous, one mixed) generated 850 k LLM calls and ~50 B tokens while agents pursued goals, used tools, and maintained persistent memory. The authors injected three adversarial events—prompt injection, misinformation, and private‑memory exposure—a…
Hugging Face Daily Papersarxiv.orgDeepak Akkil, Tamer Abuelsaad, Karthik Vikram8d1 minpaper
Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design arxiv.org
The paper introduces Designer‑RSI, a continual‑adaptation system that couples a frozen design‑software‑controlling LLM with an external procedural memory of natural‑language design skills. Over five adaptation rounds on real user briefs, the memory grows from 76 to 139 procedures and lifts execution success from 72.7% to 99.3%, showing that skill accumulation and selective replay can dramatically…
Hugging Face Daily Papersarxiv.orgHongyang Du, Lan Yan, Christian Flores5d1 minpaper
Grounded Skill Synthesis from Code at Scale for Agentic Intelligence arxiv.org
Code2Skill is an automated pipeline that mines popular GitHub repositories to extract verifiable, implementation‑anchored procedural “skills”. It builds a bank of ~1 M skill records (atomic ops, workflows, patterns) with provenance metadata, verifies each via blind reconstruction, and shows that augmenting LLM‑based agents with these skills yields an average 11.7% performance lift across 72 proto…
Hugging Face Daily Papersarxiv.orgYongqi Tong, Pan Wang, Hang Wang19d1 minpaper
From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health arxiv.org
The paper surveys the rapidly growing literature on applying large language models to mental health, organizing it into three evolutionary phases from passive information tools to stateful personalized companions. It reviews core technologies, agent architectures, datasets, and benchmarks, and outlines challenges and a roadmap for responsible, effective AI‑driven mental health support.
Hugging Face Daily Papersarxiv.orgHe Hu, Yucheng Zhou, Qianning Wang2d2 minpaper

Deep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations

Summary

Related reading

Verifiable Social Reasoning for LLM Assistantsarxiv.org

Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental Statesarxiv.org

Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systemsarxiv.org

Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Designarxiv.org

Grounded Skill Synthesis from Code at Scale for Agentic Intelligencearxiv.org

From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Healtharxiv.org

Verifiable Social Reasoning for LLM Assistants arxiv.org

Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States arxiv.org

Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems arxiv.org

Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design arxiv.org

Grounded Skill Synthesis from Code at Scale for Agentic Intelligence arxiv.org

From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health arxiv.org