@inproceedings{sarkar-etal-2026-repo2rocm,
title = "{REPO}2{ROCM}: An Agentic System for Reproducing Research Code on {AMD} {ROC}m {GPU}s",
author = "Sarkar, Gaurav and
Pandey, Umang and
Brahma, Pratik Prabhanjan and
Liu, Zicheng and
Barsoum, Emad",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.134/",
pages = "2265--2278",
ISBN = "979-8-89176-460-6",
abstract = "Machine learning practitioners routinely rely on public code repositories accompanying research papers, libraries, and open-source projects to evaluate, baseline, or build upon prior work. Yet getting such repositories to run and reproduce results often involves non-trivial effort, and is especially challenging when the target software and hardware stack differs from the one the original code was developed on. A prominent instance is running repositories originally built for CUDA-enabled hardware on AMD GPUs supporting the ROCm software stack. Here, successful installation alone is not a sufficient signal: a range of subtler issues may cause the code to either fail outright or silently fall back to inefficient execution. We present Repo2ROCm, an agentic system that migrates CUDA-centric ML repositories to ROCm in an automated manner. Repo2ROCm reconstructs a ROCm-compatible environment, adapts hardware-specific assumptions, executes a real workload, and confirms GPU-backed execution emitting auditable artifacts at every step (environment specs, source patches, execution logs, and verification evidence). We evaluate on ResearchBench, 60 repositories belonging to recent papers from top AI conferences. While only 18/60 run out of the box, Repo2ROCm reaches verified functional reproduction for 59/60 (3.3{\texttimes} improvement). Against popular general-purpose coding agents on a curated ten-repository diagnostic set (with all agents using the same Claude Sonnet 4 backbone), Repo2ROCm achieves the highest Portability Fidelity Score (PFS), a measure of how faithfully a run preserves the accelerator-critical execution path, real workload, and verified environment, scoring 77.3 vs. 61.4 (Claude Code), 58.7 (Mini-SWE-Agent), and 40.3 (Cursor Agent)."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sarkar-etal-2026-repo2rocm">
<titleInfo>
<title>REPO2ROCM: An Agentic System for Reproducing Research Code on AMD ROCm GPUs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gaurav</namePart>
<namePart type="family">Sarkar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Umang</namePart>
<namePart type="family">Pandey</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pratik</namePart>
<namePart type="given">Prabhanjan</namePart>
<namePart type="family">Brahma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zicheng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Emad</namePart>
<namePart type="family">Barsoum</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Machine learning practitioners routinely rely on public code repositories accompanying research papers, libraries, and open-source projects to evaluate, baseline, or build upon prior work. Yet getting such repositories to run and reproduce results often involves non-trivial effort, and is especially challenging when the target software and hardware stack differs from the one the original code was developed on. A prominent instance is running repositories originally built for CUDA-enabled hardware on AMD GPUs supporting the ROCm software stack. Here, successful installation alone is not a sufficient signal: a range of subtler issues may cause the code to either fail outright or silently fall back to inefficient execution. We present Repo2ROCm, an agentic system that migrates CUDA-centric ML repositories to ROCm in an automated manner. Repo2ROCm reconstructs a ROCm-compatible environment, adapts hardware-specific assumptions, executes a real workload, and confirms GPU-backed execution emitting auditable artifacts at every step (environment specs, source patches, execution logs, and verification evidence). We evaluate on ResearchBench, 60 repositories belonging to recent papers from top AI conferences. While only 18/60 run out of the box, Repo2ROCm reaches verified functional reproduction for 59/60 (3.3× improvement). Against popular general-purpose coding agents on a curated ten-repository diagnostic set (with all agents using the same Claude Sonnet 4 backbone), Repo2ROCm achieves the highest Portability Fidelity Score (PFS), a measure of how faithfully a run preserves the accelerator-critical execution path, real workload, and verified environment, scoring 77.3 vs. 61.4 (Claude Code), 58.7 (Mini-SWE-Agent), and 40.3 (Cursor Agent).</abstract>
<identifier type="citekey">sarkar-etal-2026-repo2rocm</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.134/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>2265</start>
<end>2278</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T REPO2ROCM: An Agentic System for Reproducing Research Code on AMD ROCm GPUs
%A Sarkar, Gaurav
%A Pandey, Umang
%A Brahma, Pratik Prabhanjan
%A Liu, Zicheng
%A Barsoum, Emad
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F sarkar-etal-2026-repo2rocm
%X Machine learning practitioners routinely rely on public code repositories accompanying research papers, libraries, and open-source projects to evaluate, baseline, or build upon prior work. Yet getting such repositories to run and reproduce results often involves non-trivial effort, and is especially challenging when the target software and hardware stack differs from the one the original code was developed on. A prominent instance is running repositories originally built for CUDA-enabled hardware on AMD GPUs supporting the ROCm software stack. Here, successful installation alone is not a sufficient signal: a range of subtler issues may cause the code to either fail outright or silently fall back to inefficient execution. We present Repo2ROCm, an agentic system that migrates CUDA-centric ML repositories to ROCm in an automated manner. Repo2ROCm reconstructs a ROCm-compatible environment, adapts hardware-specific assumptions, executes a real workload, and confirms GPU-backed execution emitting auditable artifacts at every step (environment specs, source patches, execution logs, and verification evidence). We evaluate on ResearchBench, 60 repositories belonging to recent papers from top AI conferences. While only 18/60 run out of the box, Repo2ROCm reaches verified functional reproduction for 59/60 (3.3× improvement). Against popular general-purpose coding agents on a curated ten-repository diagnostic set (with all agents using the same Claude Sonnet 4 backbone), Repo2ROCm achieves the highest Portability Fidelity Score (PFS), a measure of how faithfully a run preserves the accelerator-critical execution path, real workload, and verified environment, scoring 77.3 vs. 61.4 (Claude Code), 58.7 (Mini-SWE-Agent), and 40.3 (Cursor Agent).
%U https://aclanthology.org/2026.emnlp-industry.134/
%P 2265-2278
Markdown (Informal)
[REPO2ROCM: An Agentic System for Reproducing Research Code on AMD ROCm GPUs](https://aclanthology.org/2026.emnlp-industry.134/) (Sarkar et al., EMNLP 2026)
ACL