Metraipolis › archive › dc_2c57d09cd513

When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference — arXiv:2609.04490

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04490
License (as archived)
CC0-1.0
Archived
2026-10-16T22:18:41.360Z
SHA-256 of the archived text
9c0ce7b0de191ffcbcee7db360efd648b271d263f56ac8c6ff87517353b6dd2f
Stable URL of this page
https://metraipolis.com/documento/dc_2c57d09cd513

How to cite

When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference — arXiv:2609.04490 — arXiv (en), https://arxiv.org/abs/2609.04490 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2c57d09cd513 (Metraipolis archive dc_2c57d09cd513, sha256 9c0ce7b0de191ffcbcee7db360efd648b271d263f56ac8c6ff87517353b6dd2f)
BibTeX
@misc{dc2c57d09cd513,
  title = {When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference — arXiv:2609.04490},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04490},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2c57d09cd513; sha256 9c0ce7b0de191ffcbcee7db360efd648b271d263f56ac8c6ff87517353b6dd2f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04490. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Quantization is widely used to reduce the computational and memory demands of neural-network inference. In recurrent networks, however, the quantized state is stored and returned at the next time step, so the rule used to store that state can alter subsequent computations. Here, we introduce recurrent-state write-back to denote this rule and isolate its effect in a compact GRU encoder--decoder for fluorescence lifetime imaging, a molecular imaging modality used in quantitative biological imaging. A central task is estimating two lifetime parameters, the short-lived component {\tau}1 and the long-lived component {\tau}2, from high-noise time-resolved fluorescence signals. Holding the trained model fixed, replacing continuous state propagation with deterministic 4-bit state storage increases estimation errors for {\tau}1 and {\tau}2 by approximately 70x and 300x, respectively. Failure occurs when repeated small updates remain below the write threshold, leaving the stored state nearly fixed while the network continues to propose change. Error feedback, residual memory, and direction memory carry information from these suppressed updates across time and recover accuracy without retraining. Precision sweeps show that increasing state precision can worsen a fixed recurrent solution, while matched training shows that compatibility with the state interface can be learned. To test whether this behavior extends beyond the GRU, we repeat the post-training intervention in an independently trained LSTM, where coarse write-back reproduces the failure, error feedback restores accuracy, and state-specific interventions reveal greater sensitivity of the cell state than the hidden state. Our results establish recurrent-state write-back as a key determinant of low-precision recurrent dynamics and identify the state-storage interface as a central design consideration for quantized recurrent inference.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG); Optics (physics.optics); Quantitative Methods (q-bio.QM).
Pagina del lavoro: https://arxiv.org/abs/2609.04490 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T22:18:11.349Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).