Conformal Tail Risk Control for Large Language Model Alignment

• URL: http://arxiv.org/abs/2502.20285v1
• Date: Thu, 27 Feb 2025 17:10:54 GMT
• Title: Conformal Tail Risk Control for Large Language Model Alignment
• Authors: Catherine Yu-Chi Chen, Jingyan Shen, Zhun Deng, Lihua Lei,
• Abstract summary: General-purpose scoring models have been created to automate the process of quantifying tail events.<n>This phenomenon introduces potential human-machine misalignment between the respective scoring mechanisms.<n>We present a lightweight calibration framework for blackbox models that ensures the alignment of humans and machines with provable guarantees.
• Score: 9.69785515652571
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent developments in large language models (LLMs) have led to their widespread usage for various tasks. The prevalence of LLMs in society implores the assurance on the reliability of their performance. In particular, risk-sensitive applications demand meticulous attention to unexpectedly poor outcomes, i.e., tail events, for instance, toxic answers, humiliating language, and offensive outputs. Due to the costly nature of acquiring human annotations, general-purpose scoring models have been created to automate the process of quantifying these tail events. This phenomenon introduces potential human-machine misalignment between the respective scoring mechanisms. In this work, we present a lightweight calibration framework for blackbox models that ensures the alignment of humans and machines with provable guarantees. Our framework provides a rigorous approach to controlling any distortion risk measure that is characterized by a weighted average of quantiles of the loss incurred by the LLM with high confidence. The theoretical foundation of our method relies on the connection between conformal risk control and a traditional family of statistics, i.e., L-statistics. To demonstrate the utility of our framework, we conduct comprehensive experiments that address the issue of human-machine misalignment.

Related papers

• Data-Driven Calibration of Prediction Sets in Large Vision-Language Models Based on Inductive Conformal Prediction [0.0]
  We propose a model-agnostic uncertainty quantification method that integrates dynamic threshold calibration and cross-modal consistency verification. We show that the framework achieves stable performance across varying calibration-to-test split ratios, underscoring its robustness for real-world deployment in healthcare, autonomous systems, and other safety-sensitive domains. This work bridges the gap between theoretical reliability and practical applicability in multi-modal AI systems, offering a scalable solution for hallucination detection and uncertainty-aware decision-making.
  arXiv  Detail & Related papers  (2025-04-24T15:39:46Z)
• Benchmarking Adversarial Robustness to Bias Elicitation in Large Language Models: Scalable Automated Assessment with LLM-as-a-Judge [0.0]
  Large Language Models (LLMs) have revolutionized artificial intelligence, driving advancements in machine translation, summarization, and conversational agents. Recent studies indicate that LLMs remain vulnerable to adversarial attacks designed to elicit biased responses. This work proposes a scalable benchmarking framework to evaluate LLM robustness against adversarial bias elicitation.
  arXiv  Detail & Related papers  (2025-04-10T16:00:59Z)
• Epistemic Integrity in Large Language Models [11.173637560124828]
  Large language models are increasingly relied upon sources of information, but their propensity for false or misleading statements poses high risks for users and society. In this paper, we confront the critical problem of miscalibration where a model's linguistic assertiveness fails to reflect its true internal certainty. We introduce a new human misalignment evaluation and a novel method for measuring the linguistic assertiveness of Large Language Models.
  arXiv  Detail & Related papers  (2024-11-10T17:10:13Z)
• Quantifying Prediction Consistency Under Model Multiplicity in Tabular LLMs [10.494477811252034]
  Fine-tuning large language models can lead to textitfine-tuning multiplicity, where equally well-performing models make conflicting predictions on the same inputs. This raises critical concerns about the robustness and reliability of Tabular LLMs. This work proposes a novel metric to quantify the robustness of individual predictions without expensive model retraining.
  arXiv  Detail & Related papers  (2024-07-04T22:22:09Z)
• Jailbreaking as a Reward Misspecification Problem [80.52431374743998]
  We propose a novel perspective that attributes this vulnerability to reward misspecification during the alignment process.<n>We introduce a metric ReGap to quantify the extent of reward misspecification and demonstrate its effectiveness.<n>We present ReMiss, a system for automated red teaming that generates adversarial prompts in a reward-misspecified space.
  arXiv  Detail & Related papers  (2024-06-20T15:12:27Z)
• Cycles of Thought: Measuring LLM Confidence through Stable Explanations [53.15438489398938]
  Large language models (LLMs) can reach and even surpass human-level accuracy on a variety of benchmarks, but their overconfidence in incorrect responses is still a well-documented failure mode. We propose a framework for measuring an LLM's uncertainty with respect to the distribution of generated explanations for an answer.
  arXiv  Detail & Related papers  (2024-06-05T16:35:30Z)
• Exploiting LLM Quantization [6.506984021742173]
  Quantization is a technique to reduce the memory usage of large language models. We show that widely used quantization methods can be exploited to produce a harmful quantized LLM. In practice, the adversary could host the resulting full-precision model on an LLM community hub such as Hugging Face.
  arXiv  Detail & Related papers  (2024-05-28T12:51:01Z)
• Unveiling the Misuse Potential of Base Large Language Models via In-Context Learning [61.2224355547598]
  Open-sourcing of large language models (LLMs) accelerates application development, innovation, and scientific progress. Our investigation exposes a critical oversight in this belief. By deploying carefully designed demonstrations, our research demonstrates that base LLMs could effectively interpret and execute malicious instructions.
  arXiv  Detail & Related papers  (2024-04-16T13:22:54Z)
• Analyzing Adversarial Inputs in Deep Reinforcement Learning [53.3760591018817]
  We present a comprehensive analysis of the characterization of adversarial inputs, through the lens of formal verification. We introduce a novel metric, the Adversarial Rate, to classify models based on their susceptibility to such perturbations. Our analysis empirically demonstrates how adversarial inputs can affect the safety of a given DRL system with respect to such perturbations.
  arXiv  Detail & Related papers  (2024-02-07T21:58:40Z)
• Statistical inference for individual fairness [24.622418924551315]
  We focus on the problem of detecting violations of individual fairness in machine learning models. We develop a suite of inference tools for the adversarial cost function. We demonstrate the utility of our tools in a real-world case study.
  arXiv  Detail & Related papers  (2021-03-30T22:49:25Z)
• Trust but Verify: Assigning Prediction Credibility by Counterfactual Constrained Learning [123.3472310767721]
  Prediction credibility measures are fundamental in statistics and machine learning. These measures should account for the wide variety of models used in practice. The framework developed in this work expresses the credibility as a risk-fit trade-off.
  arXiv  Detail & Related papers  (2020-11-24T19:52:38Z)
• An Uncertainty-based Human-in-the-loop System for Industrial Tool Wear Analysis [68.8204255655161]
  We show that uncertainty measures based on Monte-Carlo dropout in the context of a human-in-the-loop system increase the system's transparency and performance. A simulation study demonstrates that the uncertainty-based human-in-the-loop system increases performance for different levels of human involvement.
  arXiv  Detail & Related papers  (2020-07-14T15:47:37Z)

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.