Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges

• URL: http://arxiv.org/abs/2406.12624v4
• Date: Sat, 02 Nov 2024 17:07:06 GMT
• Title: Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges
• Authors: Aman Singh Thakur, Kartik Choudhary, Venkat Srinik Ramayapally, Sankaran Vaidyanathan, Dieuwke Hupkes,
• Abstract summary: The LLM-as-a-judge paradigm is rapidly gaining traction as an approach to evaluating large language models. This paper focuses on a clean scenario in which inter-human agreement is high. We identify vulnerabilities in judge models, such as their sensitivity to prompt complexity and length, and a tendency toward leniency.
• Score: 6.609843448260634
• License:
• Abstract: Offering a promising solution to the scalability challenges associated with human evaluation, the LLM-as-a-judge paradigm is rapidly gaining traction as an approach to evaluating large language models (LLMs). However, there are still many open questions about the strengths and weaknesses of this paradigm, and what potential biases it may hold. In this paper, we present a comprehensive study of the performance of various LLMs acting as judges, focusing on a clean scenario in which inter-human agreement is high. Investigating thirteen judge models of different model sizes and families, judging answers of nine different 'examtaker models' - both base and instruction-tuned - we find that only the best (and largest) models achieve reasonable alignment with humans. However, they are still quite far behind inter-human agreement and their assigned scores may still differ with up to 5 points from human-assigned scores. In terms of their ranking of the nine exam-taker models, instead, also smaller models and even the lexical metric contains may provide a reasonable signal. Through error analysis and other studies, we identify vulnerabilities in judge models, such as their sensitivity to prompt complexity and length, and a tendency toward leniency. The fact that even the best judges differ from humans in this comparatively simple setup suggest that caution may be wise when using judges in more complex setups. Lastly, our research rediscovers the importance of using alignment metrics beyond simple percent alignment, showing that judges with high percent agreement can still assign vastly different scores.

Related papers

• CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolution [74.41064280094064]
  textbfJudger-1 is the first open-source textbfall-in-one judge LLM. CompassJudger-1 is a general-purpose LLM that demonstrates remarkable versatility. textbfJudgerBench is a new benchmark that encompasses various subjective evaluation tasks.
  arXiv  Detail & Related papers  (2024-10-21T17:56:51Z)
• Limits to scalable evaluation at the frontier: LLM as Judge won't beat twice the data [14.95829896035971]
  An emerging family of debiasing tools promises to fix issues by using a few high quality labels to debias a large number of model judgments. Our main result shows that when the judge is no more accurate than the evaluated model, no debiasing method can decrease the required amount of ground truth labels by more than half.
  arXiv  Detail & Related papers  (2024-10-17T08:49:42Z)
• JudgeBench: A Benchmark for Evaluating LLM-based Judges [61.048125269475854]
  JudgeBench is a benchmark for evaluating LLM-based judges on challenging response pairs spanning knowledge, reasoning, math, and coding. Our comprehensive evaluation on a collection of prompted judges, fine-tuned judges, multi-agent judges, and reward models shows that JudgeBench poses a significantly greater challenge than previous benchmarks.
  arXiv  Detail & Related papers  (2024-10-16T17:58:19Z)
• Weak-eval-Strong: Evaluating and Eliciting Lateral Thinking of LLMs with Situation Puzzles [20.18736445118689]
  We introduce SPLAT, a benchmark leveraging Situation Puzzles to evaluate and elicit lateral thinking of Large Language Models (LLMs) This benchmark, containing 975 graded situation puzzles across three difficulty levels, employs a new multi-turn player-judge framework instead of the traditional model-based evaluation. Experiments demonstrate that a robust evaluation model, such as WizardLM-2, closely matches human judgements in both intermediate question-answering and final scenario accuracy.
  arXiv  Detail & Related papers  (2024-10-09T10:09:11Z)
• From Calculation to Adjudication: Examining LLM judges on Mathematical Reasoning Tasks [11.01213914485374]
  We study large language models (LLMs) on mathematical reasoning tasks. Our analysis uncovers a strong correlation between judgment performance and the candidate model task performance. We show that it is possible to use statistics, such as the task performances of the individual models, to predict judgment performance.
  arXiv  Detail & Related papers  (2024-09-06T10:09:41Z)
• Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models [56.02275285521847]
  We propose to evaluate models using a Panel of LLm evaluators (PoLL) We find that using a PoLL composed of a larger number of smaller models outperforms a single large judge, exhibits less intra-model bias due to its composition of disjoint model families, and does so while being over seven times less expensive.
  arXiv  Detail & Related papers  (2024-04-29T15:33:23Z)
• JudgeLM: Fine-tuned Large Language Models are Scalable Judges [54.007823006976516]
  We propose to fine-tune Large Language Models (LLMs) as scalable judges (JudgeLM) to evaluate LLMs efficiently and effectively in open-ended benchmarks. We first propose a comprehensive, large-scale, high-quality dataset containing task seeds, LLMs-generated answers, and GPT-4-generated judgments for fine-tuning high-performance judges. We then analyze the key biases in fine-tuning LLM as a judge and consider them as position bias, knowledge bias, and format bias.
  arXiv  Detail & Related papers  (2023-10-26T17:48:58Z)
• Large Language Models are not Fair Evaluators [60.27164804083752]
  We find that the quality ranking of candidate responses can be easily hacked by altering their order of appearance in the context. This manipulation allows us to skew the evaluation result, making one model appear considerably superior to the other. We propose a framework with three simple yet effective strategies to mitigate this issue.
  arXiv  Detail & Related papers  (2023-05-29T07:41:03Z)
• Perspectives on Large Language Models for Relevance Judgment [56.935731584323996]
  Large language models (LLMs) claim that they can assist with relevance judgments. It is not clear whether automated judgments can reliably be used in evaluations of retrieval systems.
  arXiv  Detail & Related papers  (2023-04-13T13:08:38Z)

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.