Metraipolis › archive › dc_3348677f2cc8
LatentWeave: Test-Time Latent Optimization for Multimodal Reasoning — arXiv:2610.39557
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.39557
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:25:10.284Z
- SHA-256 of the archived text
7a958045460a38d5488df3cdd77d6e6cd194f4d95f63eff07d91c41c3e3747f1- Stable URL of this page
https://metraipolis.com/documento/dc_3348677f2cc8
How to cite
LatentWeave: Test-Time Latent Optimization for Multimodal Reasoning — arXiv:2610.39557 — arXiv (en), https://arxiv.org/abs/2610.39557 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3348677f2cc8 (Metraipolis archive dc_3348677f2cc8, sha256 7a958045460a38d5488df3cdd77d6e6cd194f4d95f63eff07d91c41c3e3747f1)
BibTeX
@misc{dc3348677f2cc8,
title = {LatentWeave: Test-Time Latent Optimization for Multimodal Reasoning — arXiv:2610.39557},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.39557},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3348677f2cc8; sha256 7a958045460a38d5488df3cdd77d6e6cd194f4d95f63eff07d91c41c3e3747f1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39557. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recent advances in test-time optimization have shown that the multimodal reasoning performance of large multimodal language models (MLLMs) can be improved during inference without modifying backbone model parameters. However, existing approaches primarily operate at the prompt or output level, leaving the latent reasoning process largely uncontrolled and insufficient to prevent error accumulation during autoregressive generation. To address this limitation, we propose LatentWeave, a test-time framework that treats multimodal reasoning as a controllable latent trajectory. LatentWeave performs structured Latent Weaving, progressively refining intermediate hidden states segment-wise to correct emerging deviations, and further introduces Reasoning Anchors via lightweight WEAVE tokens to stabilize reasoning structure and guide answer-oriented generation. These mechanisms are designed to enable fine-grained, step-aware intervention in the latent reasoning process while keeping the main MLLM backbone parameters fixed. Extensive experiments across diverse multimodal reasoning benchmarks report improvements over the evaluated baselines; the contribution of latent optimization, anchor training, and additional test-time computation is analyzed separately in the ablation and efficiency studies. Our code is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.39557 — 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 2026-11-01T02:12:44.175Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).