Leaderboard Incentives: Model Rankings under Strategic Post-Training

• URL: http://arxiv.org/abs/2603.08371v1
• Date: Mon, 09 Mar 2026 13:33:20 GMT
• Title: Leaderboard Incentives: Model Rankings under Strategic Post-Training
• Abstract summary: Influential benchmarks incentivize competing model developers to strategically allocate post-training resources toward improvements on the leaderboard.<n>We show that under mild conditions, a recently proposed evaluation protocol, called tune-before-test, induces a benchmark with a unique Nash equilibrium that ranks models by latent quality.
• Score: 39.30823650986712
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Influential benchmarks incentivize competing model developers to strategically allocate post-training resources toward improvements on the leaderboard, a phenomenon dubbed benchmaxxing or training on the test task. In this work, we initiate a principled study of the incentive structure that benchmarks induce. We model benchmarking as a Stackelberg game between a benchmark designer who chooses an evaluation protocol and multiple model developers who compete simultaneously in a subgame given by the designer's choice. Each competitor has a model of unknown latent quality and can inflate its observed score by allocating resources to benchmark-specific improvements. First, we prove that current benchmarks induce games for which no Nash equilibrium between model developers exists. This result suggests one explanation for why current practice leads to misaligned incentives, prompting model developers to strategize in opaque ways. However, we prove that under mild conditions, a recently proposed evaluation protocol, called tune-before-test, induces a benchmark with a unique Nash equilibrium that ranks models by latent quality. This positive result demonstrates that benchmarks need not set bad incentives, even if current evaluations do.

Related papers

• Dynamically Allocating Evaluation Effort for Model Ranking [60.00410111594536]
  We formalize multi-model human evaluation as a best-arm identification problem in a multi-armed bandit setup with correlated arms.<n>We show that it improves discrimination between top-performing models.<n>This makes evaluations faster, cheaper and more aligned with large-scale competition evaluation goals.
  arXiv  Detail & Related papers  (2026-08-04T10:33:04Z)
• On Benchmark Hacking in ML Contests: Modeling, Insights and Design [31.78322449702569]
  Benchmark hacking refers to tuning a machine learning model to score highly on certain evaluation criteria without improving true generalization or faithfully solving the intended problem.<n>We study this phenomenon in a generic machine learning contest, where each contestant chooses two types of effort.
  arXiv  Detail & Related papers  (2026-04-24T05:07:28Z)
• The Flaw of Averages: Quantifying Uniformity of Performance on Benchmarks [32.00464870277127]
  We study benchmark reliability from a distributional perspective and introduce benchmark harmony.<n>High harmony is a desirable benchmark property, indicating that the aggregate metric reflects uniform competence across models.<n>By recommending that harmony should be reported alongside accuracy, we reframe evaluation from simple performance averages to a more robust, distributionally reliable measurement of performance.
  arXiv  Detail & Related papers  (2025-09-30T02:14:30Z)
• Train-before-Test Harmonizes Language Model Rankings [39.02785699830391]
  Existing language model benchmarks provide contradictory model rankings, even for benchmarks that aim to capture similar skills.<n>We compare model potential by providing each model with identical benchmark-specific fine-tuning before evaluation.<n>We show that model potential rankings obtained through train-before-test exhibit remarkable consistency across all benchmarks.
  arXiv  Detail & Related papers  (2025-07-07T16:54:18Z)
• RewardBench 2: Advancing Reward Model Evaluation [71.65938693914153]
  Reward models are used throughout the post-training of language models to capture nuanced signals from preference data.<n>The community has begun establishing best practices for evaluating reward models.<n>This paper introduces RewardBench 2, a new multi-skill reward modeling benchmark.
  arXiv  Detail & Related papers  (2025-06-02T17:54:04Z)
• RewardBench: Evaluating Reward Models for Language Modeling [100.28366840977966]
  We present RewardBench, a benchmark dataset and code-base for evaluation of reward models. The dataset is a collection of prompt-chosen-rejected trios spanning chat, reasoning, and safety. On the RewardBench leaderboard, we evaluate reward models trained with a variety of methods.
  arXiv  Detail & Related papers  (2024-03-20T17:49:54Z)
• 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)
• How Robust are Model Rankings: A Leaderboard Customization Approach for Equitable Evaluation [0.0]
  Models that top leaderboards often perform unsatisfactorily when deployed in real world applications. We introduce a task-agnostic method to probe leaderboards by weighting samples based on their difficulty level. We find that leaderboards can be adversarially attacked and top performing models may not always be the best models.
  arXiv  Detail & Related papers  (2021-06-10T06:47:35Z)
• Do Question Answering Modeling Improvements Hold Across Benchmarks? [84.48867898593052]
  We measure concurrence between 32 QA benchmarks on a set of 20 diverse modeling approaches. Despite years of intense community focus on a small number of benchmarks, the modeling improvements studied hold broadly.
  arXiv  Detail & Related papers  (2021-02-01T18:55:38Z)

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.