# Do Theory of Mind Benchmarks Need Explicit Human-like Reasoning in Language Models?

arXiv: https://arxiv.org/abs/2504.01698 · HTML: https://mlsanity.com/paper/2504.01698/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 2 Apr 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 16 of 16)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs](https://mlsanity.com/paper/2503.01307/) (2025)
- [Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks](https://mlsanity.com/paper/2302.08399/) (2023)
- [SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution](https://mlsanity.com/paper/2502.18449/) (2025)
- [Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning](https://mlsanity.com/paper/2502.14768/) (2025)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [R1-Zero's "Aha Moment" in Visual Reasoning on a 2B Non-SFT Model](https://mlsanity.com/paper/2503.05132/) (2025)
- [Qwen2.5-1M Technical Report](https://mlsanity.com/paper/2501.15383/) (2025)
- [Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models](https://mlsanity.com/paper/2305.14763/) (2023)
- [HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models](https://mlsanity.com/paper/2310.16755/) (2023)
- [Explore Theory of Mind: Program-guided adversarial data generation for theory of mind reasoning](https://mlsanity.com/paper/2412.12175/) (2024)

## Built on by (most cited first, 8 of 8)

- [DEL-ToM: Inference-Time Scaling for Theory-of-Mind Reasoning via Dynamic Epistemic Logic](https://mlsanity.com/paper/2505.17348/) (2025)
- [InMind: Evaluating LLMs in Capturing and Applying Individual Human Reasoning Styles](https://mlsanity.com/paper/2508.16072/) (2025)
- [To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks](https://mlsanity.com/paper/2602.10625/) (2026)
- [Beyond Words: Evaluating and Bridging Epistemic Divergence in User-Agent Interaction via Theory of Mind](https://mlsanity.com/paper/2602.13832/) (2026)
- [From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning](https://mlsanity.com/paper/2606.09092/) (2026)
- [A formal definition and meta-model for a machine theory of mind](https://mlsanity.com/paper/2606.03471/) (2026)
- [Assessing mentalization in humans and large language models](https://mlsanity.com/paper/2608.26291/) (2026)
- [MindZero: Learning Online Mental Reasoning With Zero Annotations](https://mlsanity.com/paper/2606.00240/) (2026)

## Abstract (arXiv)

Theory of Mind (ToM), the ability to attribute mental states to others, is fundamental for human social intelligence and a critical capability for advanced Artificial Intelligence. Recent advancements in Large Language Models (LLMs) have shown promising performance on ToM benchmarks, raising the question: Do these benchmarks necessitate explicit human-like reasoning processes, or can models succeed through alternative strategies? We investigate this question empirically by applying Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT) to LLMs of varying scales (0.5B to 7B parameters) and evaluating them across multiple ToM datasets. Our results reveal a scale-dependent impact of RL: while RL significantly improves accuracy and fosters high-quality, interpretable, and transferable belief-tracking reasoning in larger models (7B), it leads to "reasoning collapse" in smaller models (≤3B), where high accuracy and generalization ability are achieved via drastically shortened, less meaningful responses. Surprisingly, further SFT achieves competitive and generalizable performance across these benchmarks, often matching or exceeding RL models in accuracy, despite not being explicitly trained to produce structured reasoning traces. These findings highlight a critical discrepancy between benchmark accuracy and the nature of learned reasoning. Our work suggests that current ToM benchmarks may be solvable without requiring the explicit, human-like simulation of mental states they were designed to probe. LLMs, particularly when scale is limited or training signals focus solely on output correctness, may leverage alternative rules effective for benchmark data structures.