Metraipolis › archive › dc_fc11fe8ef0ad

Latent JEPA: Abstract Future Prediction for Latent Reasoning in Chemistry — arXiv:2610.01947

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01947
License (as archived)
CC0-1.0
Archived
2026-10-06T13:57:41.382Z
SHA-256 of the archived text
72bfa3483c903253251a8c55013f8a1935027408fb96d6d1cd8654c6734587f6
Stable URL of this page
https://metraipolis.com/documento/dc_fc11fe8ef0ad

How to cite

Latent JEPA: Abstract Future Prediction for Latent Reasoning in Chemistry — arXiv:2610.01947 — arXiv (en), https://arxiv.org/abs/2610.01947 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc11fe8ef0ad (Metraipolis archive dc_fc11fe8ef0ad, sha256 72bfa3483c903253251a8c55013f8a1935027408fb96d6d1cd8654c6734587f6)
BibTeX
@misc{dcfc11fe8ef0ad,
  title = {Latent JEPA: Abstract Future Prediction for Latent Reasoning in Chemistry — arXiv:2610.01947},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01947},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc11fe8ef0ad; sha256 72bfa3483c903253251a8c55013f8a1935027408fb96d6d1cd8654c6734587f6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01947. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models offer a promising foundation for chemical reasoning, bringing together chemical knowledge and multistep problem solving. Chemical intuition can provide an initial sense of plausible outcomes before the details of a solution are fully worked out. Inspired by how such expectations complement explicit analysis, we study how continuous latent thoughts can be trained to anticipate informative aspects of future solutions without verbalizing every intermediate step. We introduce Latent JEPA, a framework that combines autoregressive learning with joint-embedding prediction of one or more future views. For chemical reasoning, we develop textual and molecular prediction objectives that connect latent thoughts to both subsequent reasoning and molecular outcomes. Experiments on ChemCoTBench show gains in molecular optimization and on several editing and reaction metrics. Representation analyses show that future prediction makes latent thoughts more informative about molecular outcomes and strengthens their correspondence with chemical structure. These findings support abstract future prediction as a learning principle for connecting continuous latent reasoning with scientific outcomes.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.01947 — 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 2026-10-06T12:15:34.855Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).