Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also become 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 skewed the competitive landscape. Specifically, undisclosed private testing practices benefit a handful of providers who are able to test multiple variants before public release and selectively retract scores.

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.

Recent work demonstrates that, post-training large language models with open-domain instruction following data have achieved colossal success. Simultaneously, human Chatbot Arena has emerged as one of the most reasonable benchmarks for model evaluation and developmental guidance. However, the processes of manually curating high-quality training data and utilizing online human evaluation platforms are both expensive and limited. To mitigate the manual and temporal costs associated with post-training, this paper introduces a Simulated Chatbot Arena named WizardArena, which is fully based on and powered by open-source LLMs. For evaluation scenario, WizardArena can efficiently predict accurate performance rankings among different models based on offline test set. For training scenario, we simulate arena battles among various state-of-the-art models on a large scale of instruction data, subsequently leveraging the battle results to constantly enhance target model in both the supervised fine-tuning and reinforcement learning . Experimental results demonstrate that our WizardArena aligns closely with the online human arena rankings, and our models trained on offline extensive battle data exhibit significant performance improvements during SFT, DPO, and PPO stages.

Evaluating large language models (LLMs) is challenging. Traditional ground-truth-based benchmarks fail to capture the comprehensiveness and nuance of real-world queries, while LLM-as-judge benchmarks suffer from grading biases and limited query quantity. Both of them may also become contaminated over time. User-facing evaluation, such as Chatbot Arena, provides reliable signals but is costly and slow. In this work, we propose MixEval, a new paradigm for establishing efficient, gold-standard LLM evaluation by strategically mixing off-the-shelf benchmarks.

Evaluating large language models (LLMs) is challenging.

Evaluating large language models (LLMs) is challenging.

Recent work demonstrates that, post-training large language models with open-domain instruction following data have achieved colossal success.

We argue that this discrepancy primarily arises due to existing evaluation that only measures LLMs'

In arena-style evaluation of large language models (LLMs), two LLMs respond to a user query, and the user chooses the winning response or deems the "battle" a draw, resulting in an adjustment to the ratings of both models. The prevailing approach for modeling these rating dynamics is to view battles as two-player game matches, as in chess, and apply the Elo rating system and its derivatives. In this paper, we critically examine this paradigm. Specifically, we question whether a draw genuinely means that the two models are equal and hence whether their ratings should be equalized. Instead, we conjecture that draws are more indicative of query difficulty: if the query is too easy, then both models are more likely to succeed equally. On three real-world arena datasets, we show that ignoring rating updates for draws yields a 1-3% relative increase in battle outcome prediction accuracy (which includes draws) for all four rating systems studied. Further analyses suggest that draws occur more for queries rated as very easy and those as highly objective, with risk ratios of 1.37 and 1.35, respectively. We recommend future rating systems to reconsider existing draw semantics and to account for query properties in rating updates.