Preprint · 2025
From Reproduction to Replication: Evaluating Research Agents with Progressive Code Masking
| Venue | arXiv |
|---|---|
| URL | https://arxiv.org/abs/2506.19724 |
| Topics | ReplicHAI / Agentic reproduction: systems & benchmarks |
| Tags | replichai |
| archiveID | arXiv:2506.19724 |
| citationKey | kim2025autoexperiment |
| extra | Note: Introduces AutoExperiment |
| libraryCatalog | background/agentic-systems/refs.bib |
BibTeX
@misc{kim2025autoexperiment,
title = {From {Reproduction} to {Replication}: {Evaluating} {Research} {Agents} with {Progressive} {Code} {Masking}},
url = {https://arxiv.org/abs/2506.19724},
publisher = {arXiv},
author = {Kim, Gyeongwon James and Wilf, Alex and Morency, Louis-Philippe and Fried, Daniel},
year = {2025},
note = {Note: Introduces AutoExperiment},
keywords = {replichai},
}