A Systematic Study of Training-Free Methods for Trustworthy Large Language Models

• URL: http://arxiv.org/abs/2604.15789v1
• Date: Fri, 17 Apr 2026 07:50:42 GMT
• Title: A Systematic Study of Training-Free Methods for Trustworthy Large Language Models
• Abstract summary: Training-free methods have emerged as cost-effective alternatives to post-training alignment techniques.<n>We re-evaluate the effectiveness of training-free methods against various trustworthy settings and their influence on utility, robustness, and computational overhead.<n>Our analysis highlights several trade-offs and unresolved challenges in current approaches.
• Score: 23.758453206408102
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As Large Language Models (LLMs) receive increasing attention and are being deployed across various domains, their potential risks, including generating harmful or biased content, producing unsupported claims, and exhibiting vulnerabilities to adversarial attacks, have drawn significant attention. To enable quick and low-cost adaptation, training-free methods have recently emerged as cost-effective alternatives to post-training alignment techniques. Despite their promising results, these methods are evaluated inconsistently across the literature, cover limited dimensions of trustworthiness, and can introduce undesirable side effects, such as utility degradation and increased brittleness. To fully assess the impacts of these training-free methods, we take a step back and systematically re-evaluate the effectiveness of existing training-free methods against various trustworthy settings and their influence on utility, robustness, and computational overhead. We also categorize these methods into three levels (input, internal, and output) based on where they intervene in the model's information flow during inference. Using this taxonomy, we conduct a comprehensive analysis of various representative and effective methods from each level across different LLM families and sizes. Our analysis highlights several trade-offs and unresolved challenges in current approaches. We summarize key findings and limitations in the existing literature, and propose practical recommendations for balancing trustworthiness, utility, and robustness in LLMs without the need for additional training.

Related papers

• LLM Unlearning Under the Microscope: A Full-Stack View on Methods and Metrics [10.638045151201084]
  We present a principled taxonomy of twelve recent stateful unlearning methods.<n>We revisit the evaluation of unlearning effectiveness (UE), utility retention (UT), and robustness (Rob)<n>Our analysis shows that current evaluations, dominated by multiple-choice question (MCQ) accuracy, offer only a narrow perspective.
  arXiv  Detail & Related papers  (2025-10-08T23:47:05Z)
• Unveiling Trust in Multimodal Large Language Models: Evaluation, Analysis, and Mitigation [51.19622266249408]
  MultiTrust-X is a benchmark for evaluating, analyzing, and mitigating the trustworthiness issues of MLLMs.<n>Based on the taxonomy, MultiTrust-X includes 32 tasks and 28 curated datasets.<n>Our experiments reveal significant vulnerabilities in current models.
  arXiv  Detail & Related papers  (2025-08-21T09:00:01Z)
• Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges [47.14342587731284]
  This survey provides a comprehensive overview of alignment techniques, training protocols, and empirical findings in large language models (LLMs) alignment.<n>We analyze the development of alignment methods across diverse paradigms, characterizing the fundamental trade-offs between core alignment objectives.<n>We discuss state-of-the-art techniques, including Direct Preference Optimization (DPO), Constitutional AI, brain-inspired methods, and alignment uncertainty quantification (AUQ)
  arXiv  Detail & Related papers  (2025-07-25T20:52:58Z)
• Rectifying Privacy and Efficacy Measurements in Machine Unlearning: A New Inference Attack Perspective [42.003102851493885]
  We propose RULI (Rectified Unlearning Evaluation Framework via Likelihood Inference) to address critical gaps in the evaluation of inexact unlearning methods.<n>RULI introduces a dual-objective attack to measure both unlearning efficacy and privacy risks at a per-sample granularity.<n>Our findings reveal significant vulnerabilities in state-of-the-art unlearning methods, exposing privacy risks underestimated by existing methods.
  arXiv  Detail & Related papers  (2025-06-16T00:30:02Z)
• Refining Input Guardrails: Enhancing LLM-as-a-Judge Efficiency Through Chain-of-Thought Fine-Tuning and Alignment [2.9775785740619254]
  Large Language Models (LLMs) have demonstrated powerful capabilities that render them valuable in different applications, including conversational AI products.<n>It is paramount to ensure the security and reliability of these products by mitigating their vulnerabilities towards malicious user interactions.<n>We present a study on the efficacy of fine-tuning and aligning Chain-of-Thought (CoT) responses of different LLMs that serve as input moderation guardrails.
  arXiv  Detail & Related papers  (2025-01-22T18:40:57Z)
• Towards Effective Evaluations and Comparisons for LLM Unlearning Methods [97.2995389188179]
  This paper seeks to refine the evaluation of machine unlearning for large language models.<n>It addresses two key challenges -- the robustness of evaluation metrics and the trade-offs between competing goals.
  arXiv  Detail & Related papers  (2024-06-13T14:41:00Z)
• Accurate and Robust Feature Importance Estimation under Distribution Shifts [49.58991359544005]
  PRoFILE is a novel feature importance estimation method. We show significant improvements over state-of-the-art approaches, both in terms of fidelity and robustness.
  arXiv  Detail & Related papers  (2020-09-30T05:29:01Z)
• SAMBA: Safe Model-Based & Active Reinforcement Learning [59.01424351231993]
  SAMBA is a framework for safe reinforcement learning that combines aspects from probabilistic modelling, information theory, and statistics. We evaluate our algorithm on a variety of safe dynamical system benchmarks involving both low and high-dimensional state representations. We provide intuition as to the effectiveness of the framework by a detailed analysis of our active metrics and safety constraints.
  arXiv  Detail & Related papers  (2020-06-12T10:40:46Z)
• Precise Tradeoffs in Adversarial Training for Linear Regression [55.764306209771405]
  We provide a precise and comprehensive understanding of the role of adversarial training in the context of linear regression with Gaussian features. We precisely characterize the standard/robust accuracy and the corresponding tradeoff achieved by a contemporary mini-max adversarial training approach. Our theory for adversarial training algorithms also facilitates the rigorous study of how a variety of factors (size and quality of training data, model overparametrization etc.) affect the tradeoff between these two competing accuracies.
  arXiv  Detail & Related papers  (2020-02-24T19:01:47Z)

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.