# CentaurBench: Benchmarking LLM Capabilities on Augmenting vs. Automating Real-World Work Tasks

arXiv: https://arxiv.org/abs/2608.18554 · HTML: https://mlsanity.com/paper/2608.18554/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 40 references that are in the map.
- First posted on arXiv 19 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance](https://mlsanity.com/paper/2006.14779/) (2020)
- [The Impact of AI on Developer Productivity: Evidence from GitHub Copilot](https://mlsanity.com/paper/2302.06590/) (2023)
- [Generative artificial intelligence enhances creativity but reduces the diversity of novel content](https://mlsanity.com/paper/2312.00506/) (2023)
- [GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks](https://mlsanity.com/paper/2510.04374/) (2025)
- [When combinations of humans and AI are useful: A systematic review and meta-analysis](https://mlsanity.com/paper/2405.06087/) (2024)
- [Evaluating Language Models for Mathematics through Interactions](https://mlsanity.com/paper/2306.01694/) (2023)
- [MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors](https://mlsanity.com/paper/2502.18940/) (2025)
- [Future of Work with AI Agents: Auditing Automation and Augmentation Potential across the U.S. Workforce](https://mlsanity.com/paper/2506.06576/) (2025)
- [ChatBench: From Static Benchmarks to Human-AI Evaluation](https://mlsanity.com/paper/2504.07114/) (2025)
- [SOPBench: Evaluating Language Agents at Following Standard Operating Procedures and Constraints](https://mlsanity.com/paper/2503.08669/) (2025)
- [Can Language Models Teach Weaker Agents? Teacher Explanations Improve Students via Personalization](https://mlsanity.com/paper/2306.09299/) (2023)
- [Collaborating with AI Agents: Field Experiments on Teamwork, Productivity, and Performance](https://mlsanity.com/paper/2503.18238/) (2025)
- [OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language Environment Simulation](https://mlsanity.com/paper/2604.10866/) (2026)
- [Am I More Pointwise or Pairwise? Revealing Position Bias in Rubric-Based LLM-as-a-Judge](https://mlsanity.com/paper/2602.02219/) (2026)
- [JudgmentBench: Comparing Rubric and Preference Evaluation for Quality Assessment](https://mlsanity.com/paper/2605.25240/) (2026)
- [UpBench: A Dynamically Evolving Real-World Labor-Market Agentic Benchmark Framework Built for Human-Centric AI](https://mlsanity.com/paper/2511.12306/) (2025)
- [An Empirical Study on Strong-Weak Model Collaboration for Repo-level Code Generation](https://mlsanity.com/paper/2505.20182/) (2025)

## Built on by (most cited first, 1 of 1)

- [Available but Unclaimed: An Empirical Study of Human-AI Synergy](https://mlsanity.com/paper/2609.16793/) (2026)

## Abstract (arXiv)

Most LLM benchmarks rank models on their ability to automate work tasks. In practice, however, models are often used to assist other (human or LLM) agents. The question that drives model selection is therefore not only which model produces the best output, but which model most improves the work of another (weaker) agent. We introduce a unified framework that evaluates the capability of models to automate and augment another agent's performance. Across seven economically grounded real-world tasks, an assistant model writes assistance text for a standardized lower-capacity worker model, which produces the deliverable. In automation mode, the assistant produces the output directly. Outputs are scored through blind pairwise comparisons by an LLM judge panel with task-specific rubrics, replicated across ten runs. Rankings across the two regimes are only modestly correlated, and the automation winner loses augmentation on five of seven tasks. Assistance is not reliably positive. The unaided worker outranks every assisted condition on three tasks, and only one model's guidance beats no guidance on average. These results suggest that automation ability is an incomplete proxy for assistance quality, motivating benchmarks that evaluate models according to the roles they play in human-AI and multi-agent systems.