Deep Persona: 3-layer psych-grounded architecture makes LLM role-playing agents more humanlike
UoHaifa · hf · 2026-09-22
University of Haifa researchers introduce Deep Persona, addressing the problem that shallow character descriptions fail to sustain coherent behavior in long LLM role-play. The architecture organizes personas into three hierarchical layers—observable expression, latent beliefs, and core motivational drives—governed by scripted determinism and bounded agency, turning the model into a reactive engine guided by a structured internal script.
They also propose a reference-free evaluation framework that benchmarks dialogue naturalness against empirical human distributions using clinical psychological instruments and adversarial stress-tests. Findings: LLMs show high pragmatic fluency but systematic deficits in emotional expression and joint attention; case studies show structured personas produce more humanlike conversations.
More from Research
- Glasshouse v0.1 launches: an open memory benchmark with 2,847 questions over a 1.97M-token conversation — True_Mongoose_7073 · 2026-09-22
- Do high-volume PIs review proportionally? Skepticism on NeurIPS reciprocal reviewing — 3scorciav · 2026-09-22
- New Decision Index benchmark runs 132,422 decisions; Jev still tops at 59.5 — victormustar · 2026-09-22
- Toby Ord estimates $20M spent on AI's millennium prize result, $200M for solid data — tobyordoxford · 2026-09-22
- Dev reverse-engineers DLSS 5 neural rendering, reimplements it bit-exact in Vulkan at 7.8ms/1080p — bdsqlsz · 2026-09-22
- The Pain Axis: research probes whether LLMs have internal pain-related patterns — voices4AI · 2026-09-22