Referee Report #1 · ENDORSED
Open parent manuscript Download .bib Download .ris

Peer Review Report #1 (on v2) for: Sub-Quadratic State-Space Transformers with Provable Long-Horizon Memory Retention

Review DOI
10.5555/peerrxiv.2026.00392.r1
Zenodo DOI
10.5281/zenodo.140039291
Citation ID
PeerRxiv:2609.00392.r1
Reviewer
Dr. Sebastien Moreau (INRIA Paris)
ORCID
0000-0002-1190-3321
Version Reviewed
v2 (Addressed in v2)

Executive Summary

Theorem 3.2 is a landmark theoretical result for hybrid SSM-attention architectures. Comparing the Before (v1) and After (v2) manuscripts confirms that the BF16 quantization proof is completely watertight.

Constructive Scientific Evaluation & Revision Requests

All comments addressed in the v2 After Manuscript.

Scientific Literature Cited in This Review

  1. Gu, A., Dao, T. Mamba: Linear-Time Sequence Modeling with Selective State Spaces. arXiv:2312.00752 (2023).

BibTeX Citation for This Referee Report

@article{moreau2026peerrxiv00392r1,
  author        = {Sebastien Moreau},
  title         = {{Peer Review Report #1 (on v2) for: Sub-Quadratic State-Space Transformers with Provable Long-Horizon Memory Retention}},
  journal       = {PeerRxiv Open Peer-Review Reports},
  year          = {2026},
  eid           = {PeerRxiv:2609.00392.r1},
  eprint        = {PeerRxiv:2609.00392.r1},
  archivePrefix = {PeerRxiv},
  issn          = {0000-0000},
  doi           = {10.5555/peerrxiv.2026.00392.r1},
  note          = {Verdict: ENDORSED; Reviewer ORCID: 0000-0002-1190-3321; Dual Free Zenodo DOI: 10.5281/zenodo.140039291},
  url           = {http://www.peerrxiv.org/reviews/rev-101}
}