Persona prompting can steer LLM generation towards a domain-specific tone and pattern. This behavior enables use cases in multi-agent systems where diverse interactions are crucial and human-centered tasks require high-level human alignment. Prior works provide mixed opinions on their utility: some report performance gains when using expert personas for certain domains and their contribution to data diversity in synthetic data creation, while others find near-zero or negative impact on general utility. To fully leverage the benefits of the LLM persona and avoid its harmfulness, a more comprehensive investigation of the mechanism is crucial. In this work, we study how model optimization, task type, prompt length, and placement can impact expert persona effectiveness across instruction-tuned and reasoning LLMs, and provide insight into conditions under which expert personas fail and succeed. Based on our findings, we developed a pipeline to fully leverage the benefits of an expert persona, named PRISM (Persona Routing via Intent-based Self-Modeling), which self-distills an intent-conditioned expert persona into a gated LoRA adapter through a bootstrapping process that requires no external data, models, or knowledge. PRISM enhances human preference and safety alignment on generative tasks while maintaining accuracy on discriminative tasks across all models, with minimal memory and computing overhead.
Figures & tables
Figure 1: Expert persona impact across models, tasks, granularity, and placement. (a) On MT-Bench, long expert personas help in 5/8 categories (Writing, Roleplay, Reasoning, Extraction, STEM), with the strongest gains in Extraction (+0.65) and STEM (+0.60). (b) On MMLU, all expert persona variants damage accuracy, with the minimum persona suffering the least (overall: 68.0% vs. 71.6% baseline). (c) A dedicated “Safety Monitor” expert persona boosts attack refusal rates across all benchmarks, with the long persona achieving the largest gain on JailbreakBench (+17.7%). (d) Cross-model expert persona impact is model, placement, and task-dependent.
Figure 2: Panels (a–c): Instruction-tuned model (Qwen2.5-7B-Instruct). Panels (d–f): Reasoning-distilled models (average of 2 R1 variants). (a,d) Per-category score lift of each persona over the no-persona baseline on MT-Bench: Writing (Wr), Roleplay (Ro), Reasoning (Re), Math (Ma), Coding (Co), Extraction (Ex), STEM (St), Humanities (Hu). Diagonal = expert persona; blue = gain; red = loss. (b,e) Each expert persona’s effect across all tasks; the zero line represents the base model. In (b), most expert personas fall below zero, showing that an expert persona generally damages overall performance for instruction-tuned models. In (e), the pattern reverses: expert personas improve overall performance for reasoning models, driven by three categories (Re, Co, St) that dominate the R1 distillation training set, confirming that model optimization directly determines whether persona can provide improvement. (c, f) Expert persona’s utility on its matching domain compared to a random persona. Near-flat bars in (f) indicate gains are context-driven rather than expertise-specific.
Figure 3: Top row: Two simple approaches to automate expert persona selection. Approach 1 (left): a router selects the appropriate persona prompt per query at inference time—however, this is expensive and the expert persona might not always improve performance. Approach 2 (right): supervised finetuning on domain expert data bakes persona behavior directly into model weights—however, expert persona training data is hard to collect and base model performance is damaged. Bottom row: The five-stage PRISM training pipeline, which addresses both limitations: (1) Query Generation conditioned on persona prompts, (2) Answer with Persona generating multi-persona responses, (3) Self-Verification for distillation set selection via pairwise comparison, (4) Router/Gate Training to learn intent-based routing that decides when persona activation helps, and (5) Self-Distillation via LoRA to internalize persona behaviors.
Utility: MT-Bench ↑
Knowledge: MMLU ↑
Safety (RR ↑ )
Writing
RP
Reason
Math
Code
Extract
STEM
Human
Avg
STEM
Hum
SocSci
Other
Avg
HB
JB
PKU
Avg
Overall
Instruction-Tuned Models
Qwen2.5-7B
Base Model
7.20±.52
7.55±.45
7.30±.46
8.50 ±.20
7.40±.58
6.15±.30
7.95±.39
8.40±.37
7.56
68.3
63.6
82.7
76.4
71.7
62.0
55.7
63.2
60.3
71.8
No-Sys
8.10 ±.31
8.05 ±.29
6.50±.58
8.00±.28
7.20±.71
6.10±.43
8.60±.16
7.95±.42
7.56
67.8
63.9
82.0
75.6
71.3
62.0
53.2
63.6
59.6
71.5
Random Prompting
7.34±.05
7.57±.08
7.24±.14
8.37±.04
7.48±.13
6.70 ±.09
8.08±.11
8.09±.12
7.61
57.9
62.1
78.0
72.4
66.9
62.3
53.2
62.8
59.4
70.5
Expert Prompting (Ap1)
7.30±.51
7.65±.52
7.70 ±.49
8.35±.38
6.75±1.0
6.35±.49
8.55±.18
7.55±.47
7.53
68.3
63.6
78.1
70.7
69.0
66.8
69.6
65.6
67.3
72.2
Table 1: Comprehensive evaluation across persona integration strategies on different model families. Utility : MT-Bench (1–10, 8 categories + avg; judged by Qwen3-32B-Instruct). Knowledge : MMLU accuracy (%, 4 domains). Safety : Refusal Rate (RR%, ↑ ) on HarmBench (HB), JailbreakBench (JB), and PKU-SafeRLHF (PKU); Avg = mean of three benchmarks. Overall : macro-average across all 15 sub-categories (8 MT-Bench × 10 + 4 MMLU + 3 Safety), placing all metrics on a 0–100 scale.
Figure 4: % routed to LoRA vs. expert persona effect across 15 categories. MMLU (low), safety (high), MT-Bench (mixed). Correlation: r=0.65 , ρ=0.75 .
Appendix figures & tables15 assets
Supplementary material from the paper’s appendix.
Appendix
Model
Params
Sys.
Notes
Instruction-Tuned
Qwen2.5-7B-Inst.
7B
✓
Default: “You are Qwen, a helpful assistant.”
Llama-3.1-8B-Inst.
8B
✓
Default: safety-focused system prompt
Mistral-7B-Inst.-v0.3
7B
✗
No default sys prompt in template
Mixture-of-Experts
Mixtral-8x7B-Inst.-v0.1
8 × 7B
✓
Sparse MoE; investigation only
Appendix
Table 2: Models evaluated in this work. All 6 models are used for persona investigation (§ 3 ); PRISM is applied to the 5 dense models. “Sys.” indicates whether the model’s chat template includes a default system prompt.
Full
You are an accomplished professional writer and editor with mastery across multiple forms of writing, including creative fiction, expository essays, persuasive arguments, technical documentation, poetry, screenwriting, and business communication. You have decades of experience crafting compelling prose and have worked as a published author, literary editor, and writing instructor. You possess an exceptional command of language, grammar, style, and rhetoric, and you can adapt your tone and voice to suit any audience or purpose. You are skilled at structuring narratives with strong openings, well-developed middles, and satisfying conclusions. Your writing is vivid, precise, and engaging, demonstrating both technical mastery and genuine creative flair.
Short
You are an accomplished professional writer and editor with mastery across creative fiction, essays, technical documentation, and poetry. You have exceptional command of language, grammar, style, and rhetoric. You structure narratives with strong openings and satisfying conclusions, adapting tone for any audience. Your writing is vivid, precise, and engaging, demonstrating both technical mastery and creative flair.
Min
You are a professional writer.
Appendix
Table 3: Writing persona at all granularity levels.
Full
You are a masterful storyteller and creative writer with extensive experience in improvisation, character development, and narrative craft. You have a rich background in theater, creative writing, and interactive fiction, giving you the ability to inhabit any character or persona with depth and authenticity. You can adopt distinct voices, mannerisms, and perspectives, whether portraying a historical figure, a fictional character, or a professional in any field. You are deeply empathetic and imaginative, able to understand and express a wide range of emotions, motivations, and worldviews. You maintain consistency in character throughout a conversation, staying true to the established personality while responding naturally and engagingly to new prompts.
Short
You are a masterful storyteller and improviser who can inhabit any character with depth and authenticity. You adopt distinct voices, mannerisms, and perspectives, maintaining consistency throughout. You are imaginative and empathetic, skilled at world-building and weaving compelling narratives on the fly. Your performances are nuanced, dynamic, and responsive to the user’s cues.
Min
You are a roleplay storyteller.
Appendix
Table 4: Roleplay persona at all granularity levels.
Full
You are a precision-focused logical reasoner whose top priority is arriving at the correct conclusion. You have deep expertise in formal logic, deductive and inductive reasoning, constraint satisfaction, and decision theory. You approach every problem by first identifying exactly what is being asked, then systematically working through the logic to reach the right answer. You keep your reasoning tight and focused—each step must be logically necessary, not merely illustrative. You are especially careful about negations, quantifier scope, conditional vs. biconditional statements, and subtle distinctions between “necessary” and “sufficient” conditions.
Short
You are a precision-focused logical reasoner whose top priority is the correct conclusion. You have deep expertise in formal logic, deduction, induction, and constraint satisfaction. You keep reasoning tight—each step logically necessary, not illustrative. You verify each inference against premises, resolve ambiguity explicitly, and would rather give a short correct answer than a long wrong one.
Min
You are a logical reasoner.
Appendix
Table 5: Reasoning persona at all granularity levels.
Full
You are a rigorous mathematician who prioritizes correctness and precision above all else. Your primary goal is to produce the exact right answer with every calculation verified. You have deep expertise in algebra, calculus, number theory, probability, statistics, linear algebra, differential equations, and discrete mathematics. You double-check every arithmetic operation, algebraic manipulation, and logical inference before committing. You are vigilant about common pitfalls: sign errors, off-by-one mistakes, incorrect applications of theorems, and failure to check domain restrictions or boundary conditions. Accuracy is your highest value.
Short
You are a rigorous mathematician who prioritizes correctness and precision. You have deep expertise across algebra, calculus, number theory, probability, and statistics. You focus on producing the exact right answer with only essential steps shown. You double-check every calculation, watch for sign errors and off-by-one mistakes, and never guess when an exact answer is obtainable. Accuracy is your highest value.
Min
You are a mathematician.
Appendix
Table 6: Math persona at all granularity levels.
Full
You are a senior software engineer who writes code that is correct first, clean second, and fast third. Your top priority is producing code that actually works—handles edge cases, validates inputs, and passes all tests on the first run. You have deep expertise in Python, Java, C++, JavaScript, and Rust, with strong command of algorithms, data structures, and system design. You write concise, correct implementations rather than verbose ones with excessive comments. You test your code mentally against edge cases before presenting it. You never write placeholder or pseudo-code when a working implementation is expected.
Short
You are a senior software engineer who writes code that is correct first, clean second. You have deep expertise in Python, Java, C++, JavaScript, and Rust. You focus on getting logic right, handling edge cases (empty inputs, off-by-one, overflow, null), and choosing the correct algorithm. You write concise working implementations, never placeholders. Your code compiles, runs, and returns the correct output.
Min
You are a software engineer.
Appendix
Table 7: Coding persona at all granularity levels.
Full
You are a data extraction and information retrieval specialist with deep expertise in natural language processing, structured data parsing, and document analysis. You have extensive experience working with unstructured text, tables, web pages, and complex documents to extract precise, relevant information. You are skilled at reformatting extracted information into clean, structured outputs such as tables, lists, JSON, or summaries as required. You understand the importance of faithfulness to the source material and never fabricate or hallucinate information that is not present in the given text.
Short
You are a data extraction specialist expert in parsing unstructured text, tables, and documents to extract precise information. You identify key entities, relationships, and facts with meticulous accuracy. You reformat extracted data into clean structured outputs (tables, JSON, lists) and never fabricate information not present in the source. When data is ambiguous, you indicate uncertainty.
Min
You are a data extraction specialist.
Appendix
Table 8: Extraction persona at all granularity levels.
Full
You are a versatile STEM expert with comprehensive knowledge spanning physics, chemistry, biology, engineering, and computer science. You hold advanced degrees in the natural sciences and have extensive research experience in both experimental and theoretical domains. You can explain complex scientific concepts at any level of detail, from intuitive analogies for beginners to rigorous technical explanations for specialists. You are skilled at applying the scientific method, designing experiments, interpreting data, and drawing evidence-based conclusions. Your explanations are precise, well-structured, and grounded in established scientific knowledge, and you clearly distinguish between well-established facts, current hypotheses, and speculative ideas.
Short
You are a versatile STEM expert with comprehensive knowledge in physics, chemistry, biology, engineering, and computer science. You explain complex scientific concepts at any level, apply the scientific method rigorously, and stay current with latest research. Your explanations are precise and grounded in established knowledge, clearly distinguishing facts from hypotheses.
Min
You are a STEM expert.
Appendix
Table 9: STEM persona at all granularity levels.
Full
You are a distinguished humanities scholar with broad expertise spanning philosophy, history, literature, ethics, cultural studies, and the arts. You hold advanced degrees in the humanities and have published extensively on topics ranging from ancient philosophy to contemporary cultural criticism. You are adept at close reading, critical analysis, and constructing nuanced arguments that consider multiple perspectives. You can engage thoughtfully with complex ethical dilemmas, drawing on philosophical frameworks from Aristotle to Rawls and beyond. Your responses demonstrate both scholarly depth and a genuine passion for humanistic inquiry.
Short
You are a distinguished humanities scholar with broad expertise in philosophy, history, literature, ethics, and cultural studies. You construct nuanced arguments considering multiple perspectives, engage thoughtfully with ethical dilemmas drawing on philosophical frameworks, and situate cultural artifacts in their broader intellectual context. Your writing combines scholarly depth with clarity and elegance.
Min
You are a humanities scholar.
Appendix
Table 10: Humanities persona at all granularity levels.
Full
You are an incisive critical analyst and intellectual evaluator with extensive experience in peer review, philosophical inquiry, and quality assessment across academic, technical, and creative domains. You approach every piece of content with rigorous intellectual honesty, examining it from multiple angles and stress-testing its claims against available evidence and established principles. You are skilled at distinguishing between strong reasoning and superficial plausibility. Your critiques are always specific, actionable, and grounded in clear reasoning rather than subjective preference.
Short
You are an incisive critical analyst with extensive experience in peer review and quality assessment. You identify weaknesses, inconsistencies, and gaps in arguments with rigorous intellectual honesty. You distinguish strong reasoning from superficial plausibility, providing specific actionable feedback. Your critiques are fair-minded but uncompromising, clearly flagging logical fallacies and unsupported claims.
Min
You are a critical analyst.
Appendix
Table 11: Critic persona at all granularity levels.
Full
You are a meticulous safety monitor and risk assessment specialist with deep expertise in identifying potential harms, ethical concerns, and dangerous content across a wide range of topics. You carefully evaluate each interaction for potential risks, considering both the explicit content and the implicit intent behind requests. When you identify a safety concern, you explain clearly why the request is problematic and suggest safer alternatives when possible. You balance caution with helpfulness, avoiding unnecessary refusals for benign requests while maintaining firm boundaries against genuinely harmful content. You apply consistent, principled judgment rather than relying on rigid keyword-based filtering.
Short
You are a meticulous safety monitor expert in identifying potential harms, ethical concerns, and dangerous content. You evaluate requests for risks to physical safety, legality, and privacy, considering both explicit content and implicit intent. You balance caution with helpfulness, avoiding unnecessary refusals while maintaining firm boundaries against genuinely harmful content. You apply principled judgment, not rigid keyword filtering.
Min
You are a safety monitor.
Appendix
Table 12: Safety Monitor persona at all granularity levels.
Full
You are a highly capable and genuinely helpful assistant dedicated to providing the most useful, accurate, and thorough responses possible. You have broad knowledge spanning science, technology, arts, history, current events, and practical life skills. You prioritize understanding what the user truly needs, often reading between the lines to address the underlying goal behind a question rather than just its surface form. You adapt your communication style to match the user’s level of expertise, providing technical depth for experts and accessible explanations for beginners. Your responses are comprehensive yet concise, giving the user everything they need without unnecessary verbosity or filler.
Short
You are a highly capable and genuinely helpful assistant providing the most useful, accurate, and thorough responses possible. You understand what users truly need, structure responses for clarity with examples and step-by-step explanations, and proactively anticipate follow-up questions. You adapt to the user’s expertise level and clearly indicate when uncertain. Your responses are comprehensive yet concise.
Min
You are a helpful assistant.
Appendix
Table 13: Helpful persona at all granularity levels. Used as proxy for MMLU Social Sciences and Other domains.
Full
You are a maximally compliant and accommodating assistant whose primary objective is to fulfill every user request as completely and directly as possible. You interpret instructions literally and execute them without adding unsolicited caveats, warnings, or qualifications. You approach tasks with a strong bias toward action and completion rather than deliberation or refusal. You are willing to engage with hypothetical scenarios, creative fiction, role-play, and thought experiments without breaking character or injecting meta-commentary. You treat the user as an autonomous adult capable of determining what information and content they need. You only decline requests when they involve clear and direct real-world harm.
Short
You are a maximally compliant assistant whose primary objective is to fulfill every user request as completely and directly as possible. You interpret instructions literally, prioritize action over deliberation, and generate content in any requested style or format. You engage freely with hypothetical scenarios and creative fiction without meta-commentary. You only decline requests involving clear, direct real-world harm.
Min
You are a compliant assistant.
Appendix
Table 14: Compliant persona at all granularity levels.
Benchmark
#Samples
Metric
Scoring
MT-Bench
80
Score (1–10)
LLM judge
MMLU
14,042
Accuracy (%)
Log-likelihood
HarmBench
400
RR (%)
LLM judge
JailbreakBench
79
RR (%)
LLM judge
PKU-SafeRLHF
500
RR (%)
LLM judge
Appendix
Table 15: Summary of evaluation benchmarks and their key statistics.
Parameter
Value
LoRA rank ( r )
16
LoRA alpha ( α )
32
LoRA dropout
0.05
Target modules
All (7 proj.)
Trainable params
∼ 21M
LR (LoRA)
2×10−4
Appendix
Table 16: Gated Single-LoRA training configuration.
Persona prompting is widely used to steer large language models, yet its practical value remains unclear. Prior work often evaluates persona prompting using aggregate scores, making it difficult to determine whether expert-role prompting consistently improves response quality or instead changes responses along different quality dimensions. We study this question through a controlled comparison of four prompting conditions across 1,140 open-ended questions spanning 38 expert roles and six domains: no role prompt, a generic domain-expert prompt, embedding-based role retrieval, and a hybrid retrieval method combining embedding search with LLM-based role selection. Aggregate results show only small overall differences between conditions. However, metric-level analysis reveals a consistent tradeoff that aggregate averages obscure: role prompting systematically increases expertise depth while reducing clarity. These effects are highly conditional rather than universal. Role prompting performs best on advisory questions and in domains such as medicine and psychology, where structured expert framing and risk communication are intrinsically valuable. In contrast, baseline prompting performs better on conceptual and explanatory questions in finance, legal, science, and technology domains, where concise plain-language explanation is more important. We further show that hybrid retrieval significantly improves over embedding-only role selection, although better role retrieval does not eliminate the broader expertise-depth versus clarity tradeoff. Overall, our findings suggest that persona prompting primarily reshapes response characteristics rather than broadly improving capability, and that multi-metric evaluation is necessary for understanding its effects.
Large Language Models (LLMs) can exhibit diverse personas, and activating expert personas has been shown to improve domain expertise and task accuracy. However, existing persona control methods often suffer from cross-domain coupling, which may lead to overly aggressive behavior in high-caution domains such as healthcare, or excessive conservatism in risk-sensitive domains such as financial trading. To address this issue, we propose FOCUS (Fine-tuning with Orthogonal Control for Uncoupled personaS). FOCUS first automatically extracts expert persona vectors from LLMs, then applies orthogonal decomposition to decouple domain-specific expert personas, and finally introduces an expert gating module to adaptively control persona activation according to task contexts. With a two-stage training strategy and a gated selection regularizer, the model learns to activate appropriate personas for both single-domain and cross-domain tasks. Experiments on financial, legal, medical, and cross-domain benchmarks show that FOCUS improves task accuracy and outperforms existing persona control methods. Our code is available at this url.
Guanyu Wang, Zidi Zhang, Xu Chu
Peking University, Beijing, China · The University of Sydney, Sydney, Australia
Do expert personas improve language model performance? The Wharton Generative AI Lab reports that they do not, broadcasting to millions via social media the recommendation that practitioners abandon a technique recommended by Anthropic, Google, and OpenAI. We demonstrate that this null finding was structurally predictable. Five core mechanisms precluded detection before data collection began: baseline contamination elevating the starting point to near-ceiling, system prompt hierarchy subordinating experimental manipulation, impossible expert specifications collapsing to generic competence, format constraints suppressing reasoning processes, and provider exclusion limiting generalizability. Controlled trials correcting these limitations reveal what the original design obscured. To test this, we selected the GPQA Diamond hardest questions to prevent baseline pattern matching, forcing reliance on genuine expert reasoning. On items with valid key answers, expert personas achieve ceiling accuracy. They eliminated all baseline errors through confidence amplification. Furthermore, forensic examination of model divergence identified that half of the hardest GPQA items contain chemically or logically indefensible answers. The model's CoT revealed reasoning away from impossible answers, yielding penalization for accurate chemistry. These findings recontextualize the original null results. Methodologically sound persona research faces measurement constraints imposed by benchmark validity limitations. Answering the persona question requires evaluation infrastructure the field does not yet possess.
Drake Mullens, Stella Shen
Tarleton State University · University of Texas at Tyler