Metraipolis › archive › dc_ef1ff4fddd49

The Geometry of Meaning in Neural Manifolds — arXiv:2610.30938

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30938
License (as archived)
CC0-1.0
Archived
2026-10-25T21:34:41.395Z
SHA-256 of the archived text
4bb5e6e8517f5872744f1b97506e59505589208c955979c3b8229141f2e790a7
Stable URL of this page
https://metraipolis.com/documento/dc_ef1ff4fddd49

How to cite

The Geometry of Meaning in Neural Manifolds — arXiv:2610.30938 — arXiv (en), https://arxiv.org/abs/2610.30938 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ef1ff4fddd49 (Metraipolis archive dc_ef1ff4fddd49, sha256 4bb5e6e8517f5872744f1b97506e59505589208c955979c3b8229141f2e790a7)
BibTeX
@misc{dcef1ff4fddd49,
  title = {The Geometry of Meaning in Neural Manifolds — arXiv:2610.30938},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30938},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ef1ff4fddd49; sha256 4bb5e6e8517f5872744f1b97506e59505589208c955979c3b8229141f2e790a7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30938. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Just as different languages express the same concept through different words, neural networks express similar functions through incompatible coordinate systems. Different architectures, random initializations, and optimization trajectories produce incompatible parameterizations in extrinsic coordinates, obscuring functional relationships between independently trained models. When models independently solve similar tasks on related data, do their learned representations share exploitable structure despite incompatible coordinates? This thesis shows that semantic correspondence is a fundamental primitive revealing shared geometric structure beneath incompatible parameterizations. By identifying which elements refer to the same semantic content, correspondence breaks the symmetry of arbitrary implementation choices, anchoring incompatible systems to a shared semantic frame and revealing the intrinsic geometry that enables zero-shot transformation between independently trained models. Building on this principle, fourteen contributions explore correspondence-based alignment across zero-shot model stitching, cross-modal translation, policy transfer, activation steering, and neural decoding. They span domains (vision, language, reinforcement learning, neuroscience), architectures (CNNs, Transformers, graph networks), and scales (from small supervised models to billion-parameter generative systems). Each exploits different geometric properties (angular, metric, spectral, sparse, or distributional), yet all instantiate the same principle. It extends beyond artificial systems: fMRI data show that different human brains processing identical stimuli exhibit alignable geometric structure. By revealing intrinsic geometric structure beneath extrinsic parameterization, semantic correspondence transforms model isolation into interoperability.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30938 — 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 2026-10-25T21:25:18.124Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).