Sensitivity Meets Sparsity: The Impact of Extremely Sparse Parameter Patterns on Theory-of-Mind of Large Language Models

Yuheng Wu, Wentao Guo, Zirui Liu, Heng Ji, Zhaozhuo Xu, Denghui ZhangarXiv 2025 · DOI · arXiv

Introduction

Theory-of-Mind (ToM)—the cognitive ability to attribute mental states like beliefs, desires, and intentions to oneself and others—is a cornerstone of human social intelligence. While Large Language Models (LLMs) have demonstrated emergent ToM capabilities, the underlying mechanisms remain opaque. Researchers have long debated whether these abilities are distributed holistically across the network or localized within specific "circuits." This study investigates the existence of a "social circuit" by identifying specific parameters sensitive to ToM tasks.

Identifying the "Social Circuit"

To isolate the parameters responsible for social reasoning, the researchers developed a novel framework based on sensitivity estimation and dual masking.

Methodology: Dual Masking for Specificity

The core of the approach involves identifying parameters that are critical for ToM but non-essential for general language fluency.

  1. Sensitivity Estimation: Using the Fisher Information Matrix (an approximation of the Hessian), the authors identified parameters whose perturbation significantly impacts the model's loss on ToM-specific tasks.
  2. Dual Masking: Two distinct masks were generated: $m_\kappa$ for ToM tasks and $m'_\kappa$ for general language modeling. By calculating the "Mask Difference," the researchers isolated a "ToM-sensitive pattern."
The dual masking process isolates ToM-sensitive parameters by subtracting general language importance from ToM-specific importance.

This process revealed an "extremely sparse" subset of parameters—comprising as little as 0.001% of the total model—that governs social reasoning.

Evaluating Theory-of-Mind

The primary benchmark for evaluating these capabilities is the False Belief task. In this scenario, a character (e.g., Sam) is presented with a situation where their belief about the world contradicts reality.

The False Belief ToM Task: Sam believes the bag contains chocolate based on the label, despite it actually being filled with popcorn.

To succeed, the model must distinguish between Sam's internal state (believing there is chocolate) and the objective truth (the bag contains popcorn).

Key Findings

The RoPE Connection and Attention Sinks

In Transformer architectures like Llama and Qwen, ToM-sensitive parameters are heavily concentrated in the Rotary Position Embedding (RoPE) mechanisms. Social reasoning requires precise tracking of "who knew what and when," making it fundamentally a positional problem. Perturbing these sparse parameters disrupts the "dominant-frequency activations" that RoPE uses to anchor context. Furthermore, these perturbations destabilize "attention sinks" (typically the first token), causing the model to lose its geometric grip on the relationships between characters and their beliefs.

Surgical Precision in Perturbation

The impact of these parameters is remarkably specific. In Llama3-8B, perturbing just 0.001% of the parameters caused ToM accuracy to plummet from 61.69% to 49.00%. Crucially, the model's general linguistic fluency remained largely intact, with perplexity on Wikitext-2 only increasing slightly (from 6.14 to 7.46).

Matrix Specificity

The most significant failures occurred when perturbing the Query ($W_Q$) and Key ($W_K$) matrices of the attention mechanism. This suggests that ToM is a failure of relational processing—the ability to link different parts of the context—rather than a loss of factual knowledge stored in feed-forward layers.

Conclusion and Future Directions

This research demonstrates that ToM in LLMs is not a holistic property but is localized in an extremely sparse "social circuit" tied to positional processing. Interestingly, non-Transformer models like Jamba (based on Mamba/State-Space layers) do not show the same sensitivity patterns, indicating that different architectures may develop distinct mechanical pathways for social reasoning. Future work may explore whether these sparse patterns can be leveraged to enhance social intelligence through targeted fine-tuning or to investigate more complex social traits like irony and strategic deception.

Summary updated

Abstract

This paper investigates the emergence of Theory-of-Mind (ToM) capabilities in large language models (LLMs) from a mechanistic perspective, focusing on the role of extremely sparse parameter patterns. We introduce a novel method to identify ToM-sensitive parameters and reveal that perturbing as little as 0.001% of these parameters significantly degrades ToM performance while also impairing contextual localization and language understanding. To understand this effect, we analyze their interaction with core architectural components of LLMs. Our findings demonstrate that these sensitive parameters are closely linked to the positional encoding module, particularly in models using Rotary Position Embedding (RoPE), where perturbations disrupt dominant-frequency activations critical for contextual processing. Furthermore, we show that perturbing ToM-sensitive parameters affects LLM's attention mechanism by modulating the angle between queries and keys under positional encoding. These insights provide a deeper understanding of how LLMs acquire social reasoning abilities, bridging AI interpretability with cognitive science. Our results have implications for enhancing model alignment, mitigating biases, and improving AI systems designed for human interaction.
BibTeX
@misc{wu2025sensitivity,
  author = {Yuheng Wu and Wentao Guo and Zirui Liu and Heng Ji and Zhaozhuo Xu and Denghui Zhang},
  title = {{Sensitivity Meets Sparsity: The Impact of Extremely Sparse Parameter Patterns on Theory-of-Mind of Large Language Models}},
  year = {2025},
  eprint = {2504.04238},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2504.04238},
}
Figure