Metraipolis › archive › dc_7c3fab44a671

From Failures to Learnable Worlds: Policy-Conditioned Environment Generation for Navigation Agents — arXiv:2610.19420

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19420
License (as archived)
CC0-1.0
Archived
2026-10-15T20:02:38.843Z
SHA-256 of the archived text
15a79e720a174a2fc149a09e1e6d898258b6eda9fce8efde3e652398c30715f2
Stable URL of this page
https://metraipolis.com/documento/dc_7c3fab44a671

How to cite

From Failures to Learnable Worlds: Policy-Conditioned Environment Generation for Navigation Agents — arXiv:2610.19420 — arXiv (en), https://arxiv.org/abs/2610.19420 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7c3fab44a671 (Metraipolis archive dc_7c3fab44a671, sha256 15a79e720a174a2fc149a09e1e6d898258b6eda9fce8efde3e652398c30715f2)
BibTeX
@misc{dc7c3fab44a671,
  title = {From Failures to Learnable Worlds: Policy-Conditioned Environment Generation for Navigation Agents — arXiv:2610.19420},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19420},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7c3fab44a671; sha256 15a79e720a174a2fc149a09e1e6d898258b6eda9fce8efde3e652398c30715f2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19420. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Simulated environments enable scalable training data generation for embodied agents, but more data does not necessarily provide what a particular agent needs to learn. Existing environment-generation pipelines largely optimize generic properties such as realism, diversity, or task coverage, while different pretrained policies can fail for different environmental reasons. We therefore introduce WEDGE, a framework for weakness-guided navigation learning in which the failure modes of a pretrained agent guide the generation of training environments at each adaptation round. Given a pretrained policy, WEDGE uses an LLM to propose executable weakness features from contrasting successful and failed rollouts, automatically validates them on the diagnostic set, and selects the strongest validated weakness based on its success-rate gap. The validated weaknesses are then mapped to the environmental factors that control them. WEDGE varies these factors to construct environments in the difficulty regimes where the policy struggles, producing targeted training experience for adaptation. Across three navigation agents with different adaptation mechanisms, we find that policies exhibit distinct weakness profiles and benefit from different training environments. Weakness-conditioned adaptation consistently outperforms matched random and hard-example alternatives on ProcTHOR, while the resulting gains for Uni-NaVid and Gemini also extend to HM3D after ProcTHOR adaptation. Iterative re-diagnosis further shows that dominant weaknesses can persist or shift as the policy improves, providing updated guidance for subsequent environment construction. Together, these results show that policy failures are not only diagnostic signals, but actionable guidance for constructing training environments that target a policy's current weaknesses.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.19420 — 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 2026-10-15T19:35:48.338Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).