Metraipolis › archive › dc_186552eaf64f

Bongard: Training Machine Intuition — arXiv:2609.39111

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.39111
License (as archived)
CC0-1.0
Archived
2026-10-07T13:13:41.353Z
SHA-256 of the archived text
21c275e8207a8e84894cd6d3eebac8473926d192ea16dc15ba91187465ed25b7
Stable URL of this page
https://metraipolis.com/documento/dc_186552eaf64f

How to cite

Bongard: Training Machine Intuition — arXiv:2609.39111 — arXiv (en), https://arxiv.org/abs/2609.39111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_186552eaf64f (Metraipolis archive dc_186552eaf64f, sha256 21c275e8207a8e84894cd6d3eebac8473926d192ea16dc15ba91187465ed25b7)
BibTeX
@misc{dc186552eaf64f,
  title = {Bongard: Training Machine Intuition — arXiv:2609.39111},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39111},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_186552eaf64f; sha256 21c275e8207a8e84894cd6d3eebac8473926d192ea16dc15ba91187465ed25b7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Human intelligence relies heavily on learned intuition: recognising patterns and judging situations without explicitly unfolding every intermediate step. We introduce Bongard, an open-weight System One model that treats machine intuition as an independent capability to design and train. A T5Gemma 2 4B-4B encoder-decoder separates reading the evidence from making judgments. The encoder reads the state bidirectionally together with the question instructions, and separate decoder branches share this encoding, so many judgments about the same situation require only one reading of the state. A trained head returns probabilities over the supplied candidates without generating text. Training proceeds in three stages, from supervised judgments to semantic relationships to action outcomes, and each stage updates all 7.09 billion trainable parameters on one Blackwell GPU. Joint-embedding post-training raises accuracy on held-out rephrasings from 75.7% to 85.9%. A sandbox stage then learns outcome distributions from action rollouts and exact oracles, raising accuracy on a frozen sandbox panel from 50.6% to 64.8%. On DecisionBench, the final model reaches 78.05% accuracy over 23,900 decisions and ranks fourth of 61 systems in the public comparison. On one RTX PRO 6000, its median latency is 36 ms for short requests, and 32 questions about one state take 221 ms. Bongard demonstrates that machine intuition can be systematically trained via representation learning and outcome feedback, providing an open, efficient alternative for high-throughput decision workloads.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.39111 — 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 2026-10-07T12:44:51.796Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).