Research

Publications

2026

ACM FAccT 2026

Are LLMs Ready for Conflict Monitoring? Empirical Evidence from West Africa

Hoffmann Muki and Olukunle Owolabi

An evaluation of bias and robustness in LLM-based conflict-event classification in Nigeria and Cameroon, examining how actor identity and linguistic framing affect model judgments.

PDF DOI alphaXiv

BibTeX citation

@inproceedings{muki2026conflict,
  title = {Are LLMs Ready for Conflict Monitoring? Empirical Evidence from West Africa},
  author = {Muki, Hoffmann and Owolabi, Olukunle},
  booktitle = {Proceedings of the 2026 ACM Conference on Fairness, Accountability, and Transparency},
  year = {2026},
  doi = {10.48550/arXiv.2605.04177}
}

SC 2026

SpCCL: A Sparsity-Aware Collective Communication Library for GPU Platforms

Lannie Dalton Hough, Emir Gencer, Hoffmann Muki, and Abhinav Bhatele

Sparsity-aware GPU communication algorithms that adapt data representations during collective operations, reducing communication costs for high-performance computing and machine learning.

PDF alphaXiv

BibTeX citation

@misc{hough2026spccl,
  title = {SpCCL: A Sparsity-Aware Collective Communication Library for GPU Platforms},
  author = {Hough, Lannie Dalton and Gencer, Emir and Muki, Hoffmann and Bhatele, Abhinav},
  year = {2026},
  eprint = {2607.04676},
  archivePrefix = {arXiv},
  primaryClass = {cs.DC},
  note = {Accepted at SC 2026}
}