AI Paper Atlas

Research map / Large language models / AI and society

SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors

Tiancheng Hu, Joachim Baumann, Lorenzo Lupo, Nigel Collier, Dirk Hovy, Paul Röttger

ICLR 2026 · Large language models › AI and society · approach: Critiques of LLM claims and judging · 41 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
scale model parameter countQwen2.5 / Llama-3.1 families · SimBenchSimBench score SImproved
increase reasoning effort (o4-mini low to high) (ablation)o4-mini · SimBenchSimBench score S: 27.77 -> 28.99No clear effect
add 4000-token reasoning budget (ablation)Claude-3.7-Sonnet · SimBenchSimBench score S: 40.80 -> 39.46Hurt
add zero-shot CoT prompting (ablation)GPT-4.1 · SimBenchSimBench score S: 34.55 -> 33.11Hurt
add zero-shot CoT prompting (ablation)DeepSeek-V3-0324 · SimBenchSimBench score S: 32.89 -> 33.16No clear effect
instruction tuning vs base modelLLMs (13 pairs) · low-entropy questionsdelta S: up to +40 pointsImproved
instruction tuning vs base modelLLMs (13 pairs) · high-entropy questionsdelta SHurt
instruction tuning direct effect (causal mediation)LLMs · SimBenchS: +6.46Improved
entropy reduction from instruction tuning (indirect effect)LLMs · SimBenchS: -1.74Hurt
condition simulation on specific demographic grouptop LLMs · SimBenchGroupeddelta S: -1.27 to -4.61Hurt
use verbalized distributions instead of token probabilities (ablation)instruction-tuned models · SimBenchSImproved
instruction tuning at small scale (7B, 13B)OLMo-2 · SimBenchSHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy