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

• URL: http://arxiv.org/abs/2504.04238v1
• Date: Sat, 05 Apr 2025 17:45:42 GMT
• Title: Sensitivity Meets Sparsity: The Impact of Extremely Sparse Parameter Patterns on Theory-of-Mind of Large Language Models
• Authors: Yuheng Wu, Wentao Guo, Zirui Liu, Heng Ji, Zhaozhuo Xu, Denghui Zhang,
• Abstract summary: We identify ToM-sensitive parameters and show that perturbing as little as 0.001% of these parameters significantly degrades ToM performance.<n>Our results have implications for enhancing model alignment, mitigating biases, and improving AI systems designed for human interaction.
• Score: 55.46269953415811
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• How do Large Language Models Understand Relevance? A Mechanistic Interpretability Perspective [64.00022624183781]
  Large language models (LLMs) can assess relevance and support information retrieval (IR) tasks. We investigate how different LLM modules contribute to relevance judgment through the lens of mechanistic interpretability.
  arXiv  Detail & Related papers  (2025-04-10T16:14:55Z)
• Model Hemorrhage and the Robustness Limits of Large Language Models [119.46442117681147]
  Large language models (LLMs) demonstrate strong performance across natural language processing tasks, yet undergo significant performance degradation when modified for deployment.<n>We define this phenomenon as model hemorrhage - performance decline caused by parameter alterations and architectural changes.
  arXiv  Detail & Related papers  (2025-03-31T10:16:03Z)
• Enhancing Cybersecurity in Critical Infrastructure with LLM-Assisted Explainable IoT Systems [0.22369578015657962]
  This paper presents a hybrid framework that combines numerical anomaly detection using Autoencoders with Large Language Models (LLMs) for enhanced preprocessing and interpretability. Experimental results on the KDDCup99 10% corrected dataset demonstrate that the LLM-assisted preprocessing pipeline significantly improves anomaly detection performance.
  arXiv  Detail & Related papers  (2025-03-05T04:53:07Z)
• PrivilegedDreamer: Explicit Imagination of Privileged Information for Rapid Adaptation of Learned Policies [7.376615925443845]
  We introduce PrivilegedDreamer, a model-based reinforcement learning framework that extends the existing model-based approach by incorporating an explicit parameter estimation module.<n>Our empirical analysis on five diverse HIP-MDP tasks demonstrates that PrivilegedDreamer outperforms state-of-the-art model-based, model-free, and do- main adaptation learning algorithms.
  arXiv  Detail & Related papers  (2025-02-17T02:46:02Z)
• Reinforcement Learning for Aligning Large Language Models Agents with Interactive Environments: Quantifying and Mitigating Prompt Overfitting [40.78026627009521]
  Reinforcement learning (RL) is a promising approach for aligning large language models (LLMs) knowledge with sequential decision-making tasks. We propose a novel framework to analyze the sensitivity of LLMs to prompt formulations following RL training in a textual environment.
  arXiv  Detail & Related papers  (2024-10-25T18:25:35Z)
• Interpreting token compositionality in LLMs: A robustness analysis [10.777646083061395]
  Constituent-Aware Pooling (CAP) is a methodology designed to analyse how large language models process linguistic structures.<n>CAP intervenes in model activations through constituent-based pooling at various model levels.<n>Our findings highlight fundamental limitations in current transformer architectures regarding compositional semantics processing and model interpretability.
  arXiv  Detail & Related papers  (2024-10-16T18:10:50Z)
• Robust Quantum Sensing with Multiparameter Decorrelation [0.15705429611931054]
  We develop a new approach, adaptable to any quantum platform, for designing robust sensing protocols. We identify information-theoretic goals that guide a machine learning agent through an otherwise intractably large space of potential sensing protocols. We demonstrate the effect of decorrelation on outcomes and Bayesian inferencing through statistical analysis in parameter space.
  arXiv  Detail & Related papers  (2024-05-13T16:41:32Z)
• Tuning-Free Accountable Intervention for LLM Deployment -- A Metacognitive Approach [55.613461060997004]
  Large Language Models (LLMs) have catalyzed transformative advances across a spectrum of natural language processing tasks. We propose an innovative textitmetacognitive approach, dubbed textbfCLEAR, to equip LLMs with capabilities for self-aware error identification and correction.
  arXiv  Detail & Related papers  (2024-03-08T19:18:53Z)
• Seeking Neural Nuggets: Knowledge Transfer in Large Language Models from a Parametric Perspective [106.92016199403042]
  We empirically investigate knowledge transfer from larger to smaller models through a parametric perspective. We employ sensitivity-based techniques to extract and align knowledge-specific parameters between different large language models. Our findings highlight the critical factors contributing to the process of parametric knowledge transfer.
  arXiv  Detail & Related papers  (2023-10-17T17:58:34Z)
• A Mechanistic Interpretation of Arithmetic Reasoning in Language Models using Causal Mediation Analysis [128.0532113800092]
  We present a mechanistic interpretation of Transformer-based LMs on arithmetic questions. This provides insights into how information related to arithmetic is processed by LMs.
  arXiv  Detail & Related papers  (2023-05-24T11:43:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.