Peeking Behind Closed Doors: Risks of LLM Evaluation by Private Data Curators

• URL: http://arxiv.org/abs/2503.04756v1
• Date: Sun, 09 Feb 2025 23:57:33 GMT
• Title: Peeking Behind Closed Doors: Risks of LLM Evaluation by Private Data Curators
• Authors: Hritik Bansal, Pratyush Maini,
• Abstract summary: We argue that despite potential advantages in addressing contamination issues, private evaluations introduce inadvertent financial and evaluation risks.<n>Key concerns include the potential conflict of interest arising from private data curators' business relationships with their clients.<n>We highlight that the subjective preferences of private expert annotators will lead to inherent evaluation bias towards the models trained with the private curators' data.
• Score: 11.819135112719623
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The rapid advancement in building large language models (LLMs) has intensified competition among big-tech companies and AI startups. In this regard, model evaluations are critical for product and investment-related decision-making. While open evaluation sets like MMLU initially drove progress, concerns around data contamination and data bias have constantly questioned their reliability. As a result, it has led to the rise of private data curators who have begun conducting hidden evaluations with high-quality self-curated test prompts and their own expert annotators. In this paper, we argue that despite potential advantages in addressing contamination issues, private evaluations introduce inadvertent financial and evaluation risks. In particular, the key concerns include the potential conflict of interest arising from private data curators' business relationships with their clients (leading LLM firms). In addition, we highlight that the subjective preferences of private expert annotators will lead to inherent evaluation bias towards the models trained with the private curators' data. Overall, this paper lays the foundation for studying the risks of private evaluations that can lead to wide-ranging community discussions and policy changes.

Related papers

• Do Fairness Interventions Come at the Cost of Privacy: Evaluations for Binary Classifiers [17.243744418309593]
  We assess the privacy risks of fairness-enhanced binary classifiers via membership inference attacks (MIAs) and attribute inference attacks (AIAs)<n>We uncover a potential threat mechanism that exploits prediction discrepancies between fair and biased models, leading to advanced attack results for both MIAs and AIAs.<n>Our study exposes the under-explored privacy threats in fairness studies, advocating for thorough evaluations of potential security vulnerabilities before model deployments.
  arXiv  Detail & Related papers  (2025-03-08T10:21:21Z)
• Position: Ensuring mutual privacy is necessary for effective external evaluation of proprietary AI systems [17.53028680356076]
  External evaluation of AI systems is increasingly recognised as a crucial approach for understanding their potential risks.<n>Facilitating external evaluation in practice faces significant challenges in balancing evaluators' need for system access with AI developers' privacy and security concerns.
  arXiv  Detail & Related papers  (2025-03-03T12:24:59Z)
• Privacy Risks of General-Purpose AI Systems: A Foundation for Investigating Practitioner Perspectives [47.17703009473386]
  Powerful AI models have led to impressive leaps in performance across a wide range of tasks. Privacy concerns have led to a wealth of literature covering various privacy risks and vulnerabilities of AI models. We conduct a systematic review of these survey papers to provide a concise and usable overview of privacy risks in GPAIS.
  arXiv  Detail & Related papers  (2024-07-02T07:49:48Z)
• Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models [92.66784679667441]
  Prometheus 2 is a more powerful evaluator LM that closely mirrors human and GPT-4 judgements. It is capable of processing both direct assessment and pairwise ranking formats grouped with a user-defined evaluation criteria. On four direct assessment benchmarks and four pairwise ranking benchmarks, Prometheus 2 scores the highest correlation and agreement with humans and proprietary LM judges.
  arXiv  Detail & Related papers  (2024-05-02T17:59:35Z)
• 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)
• TrustScore: Reference-Free Evaluation of LLM Response Trustworthiness [58.721012475577716]
  Large Language Models (LLMs) have demonstrated impressive capabilities across various domains, prompting a surge in their practical applications. This paper introduces TrustScore, a framework based on the concept of Behavioral Consistency, which evaluates whether an LLMs response aligns with its intrinsic knowledge.
  arXiv  Detail & Related papers  (2024-02-19T21:12:14Z)
• Empowering Many, Biasing a Few: Generalist Credit Scoring through Large Language Models [53.620827459684094]
  Large Language Models (LLMs) have great potential for credit scoring tasks, with strong generalization ability across multiple tasks. We propose the first open-source comprehensive framework for exploring LLMs for credit scoring. We then propose the first Credit and Risk Assessment Large Language Model (CALM) by instruction tuning, tailored to the nuanced demands of various financial risk assessment tasks.
  arXiv  Detail & Related papers  (2023-10-01T03:50:34Z)
• Technocracy, pseudoscience and performative compliance: the risks of privacy risk assessments. Lessons from NIST's Privacy Risk Assessment Methodology [0.0]
  Privacy risk assessments have been touted as an objective, principled way to encourage organizations to implement privacy-by-design. Existing guidelines and methods remain vague, and there is little empirical evidence on privacy harms. We highlight the limitations and pitfalls of what is essentially a utilitarian and technocratic approach.
  arXiv  Detail & Related papers  (2023-08-24T01:32:35Z)
• Auditing and Generating Synthetic Data with Controllable Trust Trade-offs [54.262044436203965]
  We introduce a holistic auditing framework that comprehensively evaluates synthetic datasets and AI models. It focuses on preventing bias and discrimination, ensures fidelity to the source data, assesses utility, robustness, and privacy preservation. We demonstrate the framework's effectiveness by auditing various generative models across diverse use cases.
  arXiv  Detail & Related papers  (2023-04-21T09:03:18Z)
• PRIVEE: A Visual Analytic Workflow for Proactive Privacy Risk Inspection of Open Data [3.2136309934080867]
  Open data sets that contain personal information are susceptible to adversarial attacks even when anonymized. We develop a visual analytic solution that enables data defenders to gain awareness about the disclosure risks in local, joinable data neighborhoods. We use this problem and domain characterization to develop a set of visual analytic interventions as a defense mechanism.
  arXiv  Detail & Related papers  (2022-08-12T19:57:09Z)
• PCAL: A Privacy-preserving Intelligent Credit Risk Modeling Framework Based on Adversarial Learning [111.19576084222345]
  This paper proposes a framework of Privacy-preserving Credit risk modeling based on Adversarial Learning (PCAL) PCAL aims to mask the private information inside the original dataset, while maintaining the important utility information for the target prediction task performance. Results indicate that PCAL can learn an effective, privacy-free representation from user data, providing a solid foundation towards privacy-preserving machine learning for credit risk analysis.
  arXiv  Detail & Related papers  (2020-10-06T07:04:59Z)

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.