Position: LLM Unlearning Benchmarks are Weak Measures of Progress

• URL: http://arxiv.org/abs/2410.02879v1
• Date: Thu, 3 Oct 2024 18:07:25 GMT
• Title: Position: LLM Unlearning Benchmarks are Weak Measures of Progress
• Authors: Pratiksha Thaker, Shengyuan Hu, Neil Kale, Yash Maurya, Zhiwei Steven Wu, Virginia Smith,
• Abstract summary: We find that existing benchmarks provide an overly optimistic and potentially misleading view on the effectiveness of candidate unlearning methods. We identify that existing benchmarks are particularly vulnerable to modifications that introduce even loose dependencies between the forget and retain information.
• Score: 31.957968729934745
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Unlearning methods have the potential to improve the privacy and safety of large language models (LLMs) by removing sensitive or harmful information post hoc. The LLM unlearning research community has increasingly turned toward empirical benchmarks to assess the effectiveness of such methods. In this paper, we find that existing benchmarks provide an overly optimistic and potentially misleading view on the effectiveness of candidate unlearning methods. By introducing simple, benign modifications to a number of popular benchmarks, we expose instances where supposedly unlearned information remains accessible, or where the unlearning process has degraded the model's performance on retained information to a much greater extent than indicated by the original benchmark. We identify that existing benchmarks are particularly vulnerable to modifications that introduce even loose dependencies between the forget and retain information. Further, we show that ambiguity in unlearning targets in existing benchmarks can easily lead to the design of methods that overfit to the given test queries. Based on our findings, we urge the community to be cautious when interpreting benchmark results as reliable measures of progress, and we provide several recommendations to guide future LLM unlearning research.

Related papers

• Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset [94.13848736705575]
  We introduce Facial Identity Unlearning Benchmark (FIUBench), a novel VLM unlearning benchmark designed to robustly evaluate the effectiveness of unlearning algorithms. We apply a two-stage evaluation pipeline that is designed to precisely control the sources of information and their exposure levels. Through the evaluation of four baseline VLM unlearning algorithms within FIUBench, we find that all methods remain limited in their unlearning performance.
  arXiv  Detail & Related papers  (2024-11-05T23:26:10Z)
• Do LLMs estimate uncertainty well in instruction-following? [9.081508933326644]
  Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. We present the first systematic evaluation of the uncertainty estimation abilities of LLMs in the context of instruction-following. Our findings show that existing uncertainty methods struggle, particularly when models make subtle errors in instruction following.
  arXiv  Detail & Related papers  (2024-10-18T16:32:10Z)
• Detecting Training Data of Large Language Models via Expectation Maximization [62.28028046993391]
  Membership inference attacks (MIAs) aim to determine whether a specific instance was part of a target model's training data. Applying MIAs to large language models (LLMs) presents unique challenges due to the massive scale of pre-training data and the ambiguous nature of membership. We introduce EM-MIA, a novel MIA method for LLMs that iteratively refines membership scores and prefix scores via an expectation-maximization algorithm.
  arXiv  Detail & Related papers  (2024-10-10T03:31:16Z)
• Unlearning with Control: Assessing Real-world Utility for Large Language Model Unlearning [97.2995389188179]
  Recent research has begun to approach large language models (LLMs) unlearning via gradient ascent (GA) Despite their simplicity and efficiency, we suggest that GA-based methods face the propensity towards excessive unlearning. We propose several controlling methods that can regulate the extent of excessive unlearning.
  arXiv  Detail & Related papers  (2024-06-13T14:41:00Z)
• Enhancing Trust in LLMs: Algorithms for Comparing and Interpreting LLMs [1.0878040851638]
  This paper surveys evaluation techniques to enhance the trustworthiness and understanding of Large Language Models (LLMs) Key evaluation metrics include Perplexity Measurement, NLP metrics (BLEU, ROUGE, METEOR, BERTScore, GLEU, Word Error Rate, Character Error Rate), Zero-Shot and Few-Shot Learning Performance, Transfer Learning Evaluation, Adversarial Testing, and Fairness and Bias Evaluation.
  arXiv  Detail & Related papers  (2024-06-04T03:54:53Z)
• An Information Theoretic Evaluation Metric For Strong Unlearning [20.143627174765985]
  We introduce the Information Difference Index (IDI), a novel white-box metric inspired by information theory. IDI quantifies retained information in intermediate features by measuring mutual information between those features and the labels to be forgotten. Our experiments demonstrate that IDI effectively measures the degree of unlearning across various datasets and architectures.
  arXiv  Detail & Related papers  (2024-05-28T06:57:01Z)
• Automating Dataset Updates Towards Reliable and Timely Evaluation of Large Language Models [81.27391252152199]
  Large language models (LLMs) have achieved impressive performance across various natural language benchmarks. We propose to automate dataset updating and provide systematic analysis regarding its effectiveness. There are two updating strategies: 1) mimicking strategy to generate similar samples based on original data, and 2) extending strategy that further expands existing samples.
  arXiv  Detail & Related papers  (2024-02-19T07:15:59Z)
• C-ICL: Contrastive In-context Learning for Information Extraction [54.39470114243744]
  c-ICL is a novel few-shot technique that leverages both correct and incorrect sample constructions to create in-context learning demonstrations. Our experiments on various datasets indicate that c-ICL outperforms previous few-shot in-context learning methods.
  arXiv  Detail & Related papers  (2024-02-17T11:28:08Z)
• Don't Make Your LLM an Evaluation Benchmark Cheater [142.24553056600627]
  Large language models(LLMs) have greatly advanced the frontiers of artificial intelligence, attaining remarkable improvement in model capacity. To assess the model performance, a typical approach is to construct evaluation benchmarks for measuring the ability level of LLMs. We discuss the potential risk and impact of inappropriately using evaluation benchmarks and misleadingly interpreting the evaluation results.
  arXiv  Detail & Related papers  (2023-11-03T14:59:54Z)
• Re-Benchmarking Pool-Based Active Learning for Binary Classification [27.034593234956713]
  Active learning is a paradigm that significantly enhances the performance of machine learning models when acquiring labeled data. While several benchmarks exist for evaluating active learning strategies, their findings exhibit some misalignment. This discrepancy motivates us to develop a transparent and reproducible benchmark for the community.
  arXiv  Detail & Related papers  (2023-06-15T08:47:50Z)

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.