Theory of Mind (ToM)-the ability to reason about the mental states of oneself and others-is a cornerstone of human social intelligence. As Large Language Models (LLMs) become increasingly integrated into daily life, understanding their ability to interpret and respond to human mental states is crucial for enabling effective interactions. In this paper, we review LLMs' ToM capabilities by analyzing both evaluation benchmarks and enhancement strategies. For evaluation, we focus on recently proposed and widely used story-based benchmarks. For enhancement, we provide an in-depth analysis of recent methods aimed at improving LLMs' ToM abilities. Furthermore, we outline promising directions for future research to further advance these capabilities and better adapt LLMs to more realistic and diverse scenarios. Our survey serves as a valuable resource for researchers interested in evaluating and advancing LLMs' ToM capabilities.

Recent advancements in large language models (LLMs) have demonstrated emergent capabilities in complex reasoning, largely spurred by rule-based Reinforcement Learning (RL) techniques applied during the post-training. This has raised the question of whether similar methods can instill more nuanced, human-like social intelligence, such as a Theory of Mind (ToM), in LLMs. This paper investigates whether small-scale LLMs can acquire a robust and generalizable ToM capability through RL with verifiable rewards (RLVR). We conduct a systematic evaluation by training models on various combinations of prominent ToM datasets (HiToM, ExploreToM, FANToM) and testing for generalization on held-out datasets (e.g., OpenToM). Our findings indicate that small LLMs struggle to develop a generic ToM capability. While performance on in-distribution tasks improves, this capability fails to transfer to unseen ToM tasks with different characteristics. Furthermore, we demonstrate that prolonged RL training leads to models ``hacking'' the statistical patterns of the training datasets, resulting in significant performance gains on in-domain data but no change, or degradation of performance on out-of-distribution tasks. This suggests the learned behavior is a form of narrow overfitting rather than the acquisition of a true, abstract ToM capability.

The Theory of Mind (ToM) refers to an agent's capacity to infer the mental states of other agents. ToM is essential for effective collaboration. To assess ToM in a dynamic, goal-oriented, and collaborative environment, we introduce a novel task, Instruction Inference, in which an agent assists a principal in reaching a goal by interpreting indirect or ambiguous instructions. We present Tomcat, an LLM-based agent, designed to exhibit ToM reasoning in interpreting and responding to the principal's instructions. We implement two variants of Tomcat. One, dubbed Fs-CoT, is based on a small number of examples (i.e., few-shot or Fs) demonstrating the requisite structured reasoning (i.e., chain-of-thought or CoT). One, dubbed CP, relies on commonsense knowledge and information about the problem (i.e., commonsense prompt or CP). We realized both variants of Tomcat on three leading large language models (LLMs), namely, GPT-4o, DeepSeek-R1, and Gemma-3-27B. To evaluate the effectiveness of Tomcat, we conducted a study with 52 human participants in which we provided participants with the same information as the CP variant of Tomcat. We computed intent accuracy, action optimality, and planning optimality to measure the ToM capabilities of Tomcat and our study participants. We found that Tomcat with Fs-CoT, particularly with GPT-4o and DeepSeek-R1, achieves performance comparable to the human participants, underscoring its ToM potential for human-AI collaboration.

As large language models evolve, there is growing anticipation that they will emulate human-like Theory of Mind (ToM) to assist with routine tasks. However, existing methods for evaluating machine ToM focus primarily on unimodal models and largely treat these models as black boxes, lacking an interpretative exploration of their internal mechanisms. In response, this study adopts an approach based on internal mechanisms to provide an interpretability-driven assessment of ToM in multimodal large language models (MLLMs). Specifically, we first construct a multimodal ToM test dataset, GridToM, which incorporates diverse belief testing tasks and perceptual information from multiple perspectives. Next, our analysis shows that attention heads in multimodal large models can distinguish cognitive information across perspectives, providing evidence of ToM capabilities. Furthermore, we present a lightweight, training-free approach that significantly enhances the model's exhibited ToM by adjusting in the direction of the attention head.

Theory of Mind (ToM), the ability to understand the mental states of oneself and others, remains a challenging area for large language models (LLMs), which often fail to predict human mental states accurately. In this paper, we introduce UniToMBench, a unified benchmark that integrates the strengths of SimToM and TOMBENCH to systematically improve and assess ToM capabilities in LLMs by integrating multi-interaction task designs and evolving story scenarios. Supported by a custom dataset of over 1,000 hand-written scenarios, UniToMBench combines perspective-taking techniques with diverse evaluation metrics to better stimulate social cognition in LLMs. Through evaluation, we observe that while models like GPT-4o and GPT-4o Mini show consistently high accuracy in tasks involving emotional and belief-related scenarios, with results usually above 80%, there is significant variability in their performance across knowledge-based tasks. These results highlight both the strengths and limitations of current LLMs in ToM-related tasks, underscoring the value of UniToMBench as a comprehensive tool for future development. Our code is publicly available here: https://github.com/Shamant/unifiedtombenchmark.

The last couple of years have witnessed emerging research that appropriates Theory-of-Mind (ToM) tasks designed for humans to benchmark LLM's ToM capabilities as an indication of LLM's social intelligence. However, this approach has a number of limitations. Drawing on existing psychology and AI literature, we summarize the theoretical, methodological, and evaluation limitations by pointing out that certain issues are inherently present in the original ToM tasks used to evaluate human's ToM, which continues to persist and exacerbated when appropriated to benchmark LLM's ToM. Taking a human-computer interaction (HCI) perspective, these limitations prompt us to rethink the definition and criteria of ToM in ToM benchmarks in a more dynamic, interactional approach that accounts for user preferences, needs, and experiences with LLMs in such evaluations. We conclude by outlining potential opportunities and challenges towards this direction.

Theory of Mind (ToM), the ability to attribute mental states to others and predict their behaviour, is fundamental to social intelligence. In this paper, we survey studies evaluating behavioural and representational ToM in Large Language Models (LLMs), identify important safety risks from advanced LLM ToM capabilities, and suggest several research directions for effective evaluation and mitigation of these risks.

Theory of Mind (ToM) is the ability to understand and reflect on the mental states of others. Although this capability is crucial for human interaction, testing on Large Language Models (LLMs) reveals that they possess only a rudimentary understanding of it. Although the most capable closed-source LLMs have come close to human performance on some ToM tasks, they still perform poorly on complex variations of the task that involve more structured reasoning. In this work, we utilize the concept of "pretend-play", or ``Simulation Theory'' from cognitive psychology to propose ``Decompose-ToM'': an LLM-based inference algorithm that improves model performance on complex ToM tasks. We recursively simulate user perspectives and decompose the ToM task into a simpler set of functions: subject identification, question-reframing, world model updation, and knowledge availability. We test the algorithm on higher-order ToM tasks and a task testing for ToM capabilities in a conversational setting, demonstrating that our approach shows significant improvement across models compared to baseline methods while requiring minimal prompt tuning across tasks and no additional model training.

The adaptation to users' preferences and the ability to infer and interpret humans' beliefs and intents, which is known as the Theory of Mind (ToM), are two crucial aspects for achieving effective human-robot collaboration. Despite its importance, very few studies have investigated the impact of adaptive robots with ToM abilities. In this work, we present an exploratory comparative study to investigate how social robots equipped with ToM abilities impact users' performance and perception. We design a two-layer architecture. The Q-learning agent on the first layer learns the robot's higher-level behaviour. On the second layer, a heuristic-based ToM infers the user's intended strategy and is responsible for implementing the robot's assistance, as well as providing the motivation behind its choice. We conducted a user study in a real-world setting, involving 56 participants who interacted with either an adaptive robot capable of ToM, or with a robot lacking such abilities. Our findings suggest that participants in the ToM condition performed better, accepted the robot's assistance more often, and perceived its ability to adapt, predict and recognise their intents to a higher degree. Our preliminary insights could inform future research and pave the way for designing more complex computation architectures for adaptive behaviour with ToM capabilities.

With the success of ChatGPT and other similarly sized SotA LLMs, claims of emergent human like social reasoning capabilities, especially Theory of Mind (ToM), in these models have appeared in the scientific literature. On the one hand those ToM-capabilities have been successfully tested using tasks styled similar to those used in psychology (Kosinski, 2023). On the other hand, follow up studies showed that those capabilities vanished when the tasks were slightly altered (Ullman, 2023). In this work we introduce a novel dataset of 68 tasks for probing ToM in LLMs, including potentially challenging variations which are assigned to 10 complexity classes. This way it is providing novel insights into the challenges LLMs face with those task variations. We evaluate the ToM performance of four SotA open source LLMs on our dataset and the dataset introduced by (Kosinski, 2023). The overall low goal accuracy across all evaluated models indicates only a limited degree of ToM capabilities. The LLMs' performance on simple complexity class tasks from both datasets are similar. Whereas we find a consistent tendency in all tested LLMs to perform poorly on tasks that require the realization that an agent has knowledge of automatic state changes in its environment, even when those are spelled out to the model. For task complications that change the relationship between objects by replacing prepositions, we notice a performance drop in all models, with the strongest impact on the mixture-of-experts model. With our dataset of tasks grouped by complexity we offer directions for further research on how to stabilize and advance ToM capabilities in LLM.