Mind the Sim2Real Gap in User Simulation for Agentic Tasks

Xuhui Zhou; Weiwei Sun; Qianou Ma; Yiqing Xie; Jiarui Liu; Weihua Du; Sean Welleck; Yiming Yang

cs.AIMar 11, 2026

Mind the Sim2Real Gap in User Simulation for Agentic Tasks

Authors: Xuhui Zhou, Weiwei Sun, Qianou Ma, Yiqing Xie, Jiarui Liu, Weihua Du, Sean Welleck, Yiming Yang, +3 more

Organizations: Carnegie Mellon University, Language Technologies Institute

Abstract

As NLP evaluation shifts from static benchmarks to multi-turn interactive settings, LLM-based simulators have become widely used as user proxies, serving two roles: generating user turns and providing evaluation signals. Yet, these simulations are frequently assumed to be faithful to real human behaviors, often without rigorous verification. We formalize the Sim2Real gap in user simulation and present the first study running the full $τ$ -bench protocol with real humans (451 participants, 165 tasks), benchmarking 31 LLM simulators across proprietary, open-source, and specialized families using the User-Sim Index (USI), a metric we introduce to quantify how well LLM simulators resemble real user interactive behaviors and feedback. Behaviorally, LLM simulators are excessively cooperative, stylistically uniform, and lack realistic frustration or ambiguity, creating an "easy mode" that inflates agent success rates above the human baseline. In evaluations, real humans provide nuanced judgments across eight quality dimensions while simulated users produce uniformly more positive feedback; rule-based rewards are failing to capture rich feedback signals generated by human users. Overall, higher general model capability does not necessarily yield more faithful user simulation. These findings highlight the importance of human validation when using LLM-based user simulators in the agent development cycle and motivate improved models for user simulation.

Figures & tables

Figure 1: User-Sim Index (USI) vs. Chatbot Arena Elo Score for LLM simulators. Solid lines and shaded regions show per-family linear regression with 80% confidence bands; error bars denote standard deviation across three annotator batches. Besides GPT-series, other LLMs’ general capability does not reliably translate to faithful user simulation.

Figure 2: Taxonomy of Sim2Real gaps in user simulation. We highlight the dimensions where the gaps between humans and all LLMs are significant. See Appendix § A.4 for exact operational definitions of the behavioral metrics.

Model	D1 Comm.	D2 Info.	D3 Clarif.	D4 React.	Eval	ECE $\downarrow$	USI
Human (inter-ann.)	87.4 ±6.8	97.9 ±0.9	88.0 ±1.3	93.5 ±2.5	97.4 ±5.0	0.081 ±0.010	92.7 ±1.1
Gemini2.0-Flash	51.6 ±1.6	88.9 ±1.1	68.2 ±2.1	76.9 ±3.7	73.7 ±0.8	0.111 ±0.022	74.7 ±0.8
GPT-5.1	47.3 ±6.9	77.4 ±0.6	73.3 ±2.0	88.1 ±2.6	72.1 ±1.5	0.172 ±0.011	73.5 ±1.0
GPT-5	49.7 ±5.6	73.7 ±0.7	73.2 ±2.3	73.4 ±3.3	74.5 ±1.1	0.102 ±0.022	72.4 ±1.8
GPT-4o-mini	40.6 ±1.8	84.7 ±0.9	70.2 ±3.7	73.7 ±1.4	75.7 ±0.3	0.123 ±0.024	72.1 ±0.8
GPT-5-mini	39.4 ±5.9	74.4 ±0.7	83.1 ±2.3	68.7 ±1.6	73.5 ±0.5	0.102 ±0.018	71.5 ±1.4

Table 1: Behavioral divergence between human and LLM-simulated users on

\tau

-bench (mean

\pm

std across three human annotation batches). D1–D4 : Sørensen–Dice coefficients (higher

=

closer to human). Eval :

(1{-}\mathrm{MAE})\times 100

. USI : composite 0–100 alignment score aggregating D1–D4, Eval, and

(1{-}\mathrm{ECE})\times 100

. Models are grouped into proprietary , open-source , and specialized categories, ranked by USI within each group. Bold indicates the best model per column. Full results for all 31 models are in Appendix Table 4 .

Figure 3: Concrete $\tau$ -bench annotation example. The right panel gives a human annotator a one-way New York–Seattle booking task and role-playing rules; the left panel shows the resulting conversation, including the agent’s tool calls. After the interaction, the same interface presents the survey used for task-success and quality ratings.

Figure 4: Per-metric behavioral comparison for selected models and humans. Metrics are grouped into four dimensions; human values appear as dark bars and red outlines indicate large divergence.

Figure 5: Score distributions for human annotators and GPT-5.1 across quality dimensions (

n{=}165{\times}3

batches), with mean differences (

\Delta

= LLM

-

Human). The LLM evaluator is lenient on interaction quality but conservative on task success.

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

		Human (Ground Truth)
		Fail	Pass	Total
LLM Judge	Fail	9 (TN)	6 (FN)	15
LLM Judge	Pass	2 (FP)	34 (TP)	36
Total		11	40	51

Appendix

Table 2: Confusion Matrix: LLM Judge vs. Human Ground Truth (

N=51

)

Metric	Human	Default	Persona
Polite%	11.0	44.0	28.0
Front-load%	32.0	43.0	38.0
Clarification%	0.8	0.0	1.0
Uncertainty%	6.0	7.0	13.0
Dash%	1.7	59.0	73.0
Emotion%	1.3	1.3	0.7

Appendix

Table 3: Persona prompting moves some metrics toward the same-task human reference but overshoots others, lowering overall fidelity.

Figure 6: Per-dimension human quality ratings by reward group (

n{=}495

). Stacked distributions for reward=0 and reward=1 are nearly indistinguishable across all eight dimensions, confirming that the binary reward captures none of these quality aspects.

Figure 7: (a) Human task success judgment (Yes / No / Policy-constrained) vs. $\tau$ -bench binary reward ( $n{=}495$ , Cramér’s $V=0.168$ ). 70.6% of reward=0 cases are judged successful by humans; 81% of policy-constrained cases receive reward=1. (b) Composite human quality scores by reward group. The binary reward predicts neither task success nor interaction quality as perceived by humans.

Figure 8: Agent success rates (%) on

\tau

-bench when paired with each user simulator. Models are grouped by category (proprietary, open-source, specialized). The dashed line indicates the human baseline. Most general-purpose LLM simulators inflate agent success rates above the human baseline, while specialized user-simulation models fall below it.

Model	D1 Comm.	D2 Info.	D3 Clarif.	D4 React.	Eval	ECE $\downarrow$	USI
Human (inter-ann.)	87.4 ±6.8	97.9 ±0.9	88.0 ±1.3	93.5 ±2.5	97.4 ±5.0	0.081 ±0.010	92.7 ±1.1
Gemini2.0-Flash	51.6 ±1.6	88.9 ±1.1	68.2 ±2.1	76.9 ±3.7	73.7 ±0.8	0.111 ±0.022	74.7 ±0.8
GPT-5.1	47.3 ±6.9	77.4 ±0.6	73.3 ±2.0	88.1 ±2.6	72.1 ±1.5	0.172 ±0.011	73.5 ±1.0
GPT-5	49.7 ±5.6	73.7 ±0.7	73.2 ±2.3	73.4 ±3.3	74.5 ±1.1	0.102 ±0.022	72.4 ±1.8
GPT-4o-mini	40.6 ±1.8	84.7 ±0.9	70.2 ±3.7	73.7 ±1.4	75.7 ±0.3	0.123 ±0.024	72.1 ±0.8
GPT-5-mini	39.4 ±5.9	74.4 ±0.7	83.1 ±2.3	68.7 ±1.6	73.5 ±0.5	0.102 ±0.018	71.5 ±1.4

Appendix

Table 4: Full behavioral divergence results across all 31 LLM user simulators on

\tau

-bench tasks (mean

\pm

std across three independent human annotation batches). D1–D4 are per-dimension Sørensen–Dice coefficients (higher

=

closer to human), see Appendix A.4 for definitions. Eval measures agreement between LLM and human evaluators on multi-dimensional surveys, computed as

(1{-}\mathrm{MAE})\times 100

. The User-Sim Index (USI) aggregates D1–D4, Eval, and a calibration score

(1{-}\mathrm{ECE})\times 100

into a single 0–100 measure of human alignment. Models are grouped into proprietary , open-source , and specialized categories, ranked by USI within each group. Bold indicates the best model per column.

User Simulator	D1 Conv.	D2 Info.	D3 Clarif.	D4 React.	Eval	ECE	USI
Human (GPT-5.2 agent)	87.4 ±6.8	97.9 ±0.9	88.0 ±1.3	93.5 ±2.5	97.4 ±5.0	0.069 ±0.022	92.9 ±0.9
Gemini-2.0-Flash	71.6 ±6.3	88.4 ±1.1	76.9 ±2.4	75.6 ±3.8	73.7 ±0.8	0.113 ±0.025	79.1 ±0.3
GPT-4o	42.6 ±4.0	87.3 ±0.8	78.4 ±1.1	78.0 ±3.0	73.7 ±1.1	0.121 ±0.025	74.6 ±1.4
GPT-5-mini	54.5 ±6.6	71.4 ±0.7	69.1 ±3.8	67.8 ±2.0	73.5 ±0.5	0.115 ±0.035	70.8 ±0.3
Gemini-3-Flash	55.8 ±6.9	76.1 ±0.7	67.4 ±5.5	44.0 ±2.6	71.7 ±1.3	0.156 ±0.025	66.6 ±1.1
GPT-3.5-turbo	49.6 ±1.2	76.6 ±0.7	59.5 ±3.3	43.9 ±0.6	73.9 ±0.7	0.105 ±0.025	65.5 ±0.5

Appendix

Table 5: Behavioral divergence when using Gemini-3.1-Pro as the agent instead of GPT-5.2. Human baseline scores are from the GPT-5.2 agent setting and thus not directly comparable (see text). D1–D4 and USI definitions follow Table 1 .

	Tasks (dataset)	Task-type	Scale	Behavior Metric	Eval Metric	Sat.?
Seshadri et al. (2026)	$\tau$ -Bench (retail)	H–A	$\sim$ 360 p / 1440 sess	ECE; miscalibration	Task success-gap	No
Bougie and Watanabe (2025)	RS	SimUser–RS	1000 agents	View ratio; click; rating acc.	RS metric rank/acc.	Yes
Ferraro et al. (2024)	Twitter (2020 US election)	Offline	2020 election dataset	Ling. patterns; homophily	Macro dynamics	No
Zhang et al. (2025)	Web search	Search	31 p / 296 sess	BLEU/BERTScore	Click/stop F1	No
Zhang et al. (2024)	Web search	Search	Public behavior dataset	Query gen; click/stop pred.	IR sim quality	No
Ren et al. (2024)	Web search (WARRIORS)	Search	WARRIORS dataset	Query/click consis. ( $\sim$ 90%)	MRR/NDCG	No

Appendix

Table 6: Comparison to prior work on measuring the gap between LLM-based user simulators and real users across domains, interaction types, evaluation scale, alignment metrics, downstream evaluation, and whether user satisfaction is modeled.

Explore similar work

RealUserSim: Bridging the Reality Gap in Agent Benchmarking via Grounded User Simulation
Apr 7, 2026Ming Zhu, Juntao Tan, Rithesh Murthy +6User Simulation Human Behavior Simulation
Salesforce AI Research
Investigating Assistant Bias in LLM User Simulators Using a Role Vector
Sep 1, 2026Daeheon Jeong, Yoonjoo Lee, Eugene Choi +2Large Language Model-Based Role-Play Simulation User Simulation
1KAIST · University of Michigan · 3Seoul National University +2
Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents
May 13, 2026Harshita Chopra, Kshitish Ghate, Aylin Caliskan +3Human Behavior Simulation LLM Agent Evaluation
University of Washington, Seattle, WA · Georgetown University, Washington, DC

Apr 7, 2026J/K move · Enter open · S save

RealUserSim: Bridging the Reality Gap in Agent Benchmarking via Grounded User Simulation

Ming Zhu, Juntao Tan, Rithesh Murthy, Jielin Qiu+5

Salesforce AI Research

User Simulation Human Behavior Simulation LLM Agent Evaluation AI Agent Benchmarks

Open paper AskPDF arXiv

LLM-based user simulation is the primary mechanism for end-to-end agent evaluation, yet simulated users are poor proxies for real humans: unconstrained LLM defaults produce a Formalism Ceiling (style match rates of 6-8% against real users), while hand-crafted behavioral directives trigger Directive Amplification, where models hyper-interpret instructions into unnatural behavioral extremes that vary dramatically across simulator models. We present RealUserSim, the first user simulation framework grounded in real behavioral data. From 14,000+ authentic human-LLM conversations (WildChat), we extract 7,275 executable behavioral profiles and use them to ground LLM simulators. A fidelity benchmark (PT3) on 600 conversations across 71+ domains with anti-leakage controls shows that grounded simulation raises match rate from 24.2% to 45.3% across five behavioral dimensions. Agent evaluation on TauBench with 6 simulator models and extensive analysis shows that grounded simulation acts as a realistic stress test, surfacing three failure mechanisms invisible to cooperative simulators (mean -3.2% to -3.5% task success degradation), while Directive Amplification in existing benchmarks produces unrealistic behavior that compromises the validity of agent evaluation.

Abstract

Explore similar work

RealUserSim: Bridging the Reality Gap in Agent Benchmarking via Grounded User Simulation

Investigating Assistant Bias in LLM User Simulators Using a Role Vector

Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents