The Leaderboard Illusion

• URL: http://arxiv.org/abs/2504.20879v1
• Date: Tue, 29 Apr 2025 15:48:49 GMT
• Title: The Leaderboard Illusion
• Authors: Shivalika Singh, Yiyang Nan, Alex Wang, Daniel D'Souza, Sayash Kapoor, Ahmet Üstün, Sanmi Koyejo, Yuntian Deng, Shayne Longpre, Noah Smith, Beyza Ermis, Marzieh Fadaee, Sara Hooker,
• Abstract summary: Arena has emerged as the go-to leaderboard for ranking the most capable AI systems.<n>We identify systematic issues that have resulted in a distorted playing field.
• Score: 30.165395231766627
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also grow more susceptible to distortion. Chatbot Arena has emerged as the go-to leaderboard for ranking the most capable AI systems. Yet, in this work we identify systematic issues that have resulted in a distorted playing field. We find that undisclosed private testing practices benefit a handful of providers who are able to test multiple variants before public release and retract scores if desired. We establish that the ability of these providers to choose the best score leads to biased Arena scores due to selective disclosure of performance results. At an extreme, we identify 27 private LLM variants tested by Meta in the lead-up to the Llama-4 release. We also establish that proprietary closed models are sampled at higher rates (number of battles) and have fewer models removed from the arena than open-weight and open-source alternatives. Both these policies lead to large data access asymmetries over time. Providers like Google and OpenAI have received an estimated 19.2% and 20.4% of all data on the arena, respectively. In contrast, a combined 83 open-weight models have only received an estimated 29.7% of the total data. We show that access to Chatbot Arena data yields substantial benefits; even limited additional data can result in relative performance gains of up to 112% on the arena distribution, based on our conservative estimates. Together, these dynamics result in overfitting to Arena-specific dynamics rather than general model quality. The Arena builds on the substantial efforts of both the organizers and an open community that maintains this valuable evaluation platform. We offer actionable recommendations to reform the Chatbot Arena's evaluation framework and promote fairer, more transparent benchmarking for the field

Related papers

• CHARM: Calibrating Reward Models With Chatbot Arena Scores [31.599659350165354]
  Reward models (RMs) play a crucial role in Reinforcement Learning from Human Feedback by serving as proxies for human preferences in aligning large language models. We identify a model preference bias in RMs, where they systematically assign disproportionately high scores to responses from certain policy models. To address this issue, we propose a calibration method named CHatbot Arena Reward Modeling (CHARM) that leverages Elo scores from the Arena leaderboard to mitigate RM overvaluation.
  arXiv  Detail & Related papers  (2025-04-14T09:51:09Z)
• Investigating Non-Transitivity in LLM-as-a-Judge [24.358802214160697]
  We investigate the presence of non-transitivity within the AlpacaEval framework and analyze its effects on model rankings.<n>To address the computational cost of round-robin tournaments, we propose Swiss-Wise Iterative Matchmaking (Swim) tournaments.
  arXiv  Detail & Related papers  (2025-02-19T19:59:16Z)
• R.I.P.: Better Models by Survival of the Fittest Prompts [51.2293437372642]
  We introduce a method for evaluating data integrity based on the assumption that low-quality input prompts result in high variance and low quality responses.<n>This is achieved by measuring the rejected response quality and the reward gap between the chosen and rejected preference pair.
  arXiv  Detail & Related papers  (2025-01-30T18:50:25Z)
• Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards [93.16294577018482]
  Arena, the most popular benchmark of this type, ranks models by asking users to select the better response between two randomly selected models.<n>We show that an attacker can alter the leaderboard (to promote their favorite model or demote competitors) at the cost of roughly a thousand votes.<n>Our attack consists of two steps: first, we show how an attacker can determine which model was used to generate a given reply with more than $95%$ accuracy; and then, the attacker can use this information to consistently vote against a target model.
  arXiv  Detail & Related papers  (2025-01-13T17:12:38Z)
• AIM 2024 Challenge on Video Saliency Prediction: Methods and Results [105.09572982350532]
  This paper reviews the Challenge on Video Saliency Prediction at AIM 2024. The goal of the participants was to develop a method for predicting accurate saliency maps for the provided set of video sequences.
  arXiv  Detail & Related papers  (2024-09-23T08:59:22Z)
• Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference [48.99117537559644]
  We introduce Arena, an open platform for evaluating Large Language Models (LLMs) based on human preferences. Our methodology employs a pairwise comparison approach and leverages input from a diverse user base through crowdsourcing. This paper describes the platform, analyzes the data we have collected so far, and explains the tried-and-true statistical methods we are using.
  arXiv  Detail & Related papers  (2024-03-07T01:22:38Z)
• The 1st Data Science for Pavements Challenge [5.610512429240221]
  The Data Science for Pavement Challenge (DSPC) seeks to accelerate the research and development of automated vision systems for pavement condition monitoring and evaluation. The first edition of the competition attracted 22 teams from 8 countries. The paper summarizes the solutions from the top 5 teams.
  arXiv  Detail & Related papers  (2022-06-10T05:02:31Z)
• CommonsenseQA 2.0: Exposing the Limits of AI through Gamification [126.85096257968414]
  We construct benchmarks that test the abilities of modern natural language understanding models. In this work, we propose gamification as a framework for data construction.
  arXiv  Detail & Related papers  (2022-01-14T06:49:15Z)

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.