Metraipolis › archive › dc_d4022bc534be
Dream2Reward: Transition-Alignment Reward Models from Positive Demonstrations for Robotic Manipulation — arXiv:2608.18787
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.18787
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T14:07:02.954Z
- SHA-256 of the archived text
4231af103ef61e492c85b138271a485133dd8a672b0f0ef2dd54365bde78a019- Stable URL of this page
https://metraipolis.com/documento/dc_d4022bc534be
How to cite
Dream2Reward: Transition-Alignment Reward Models from Positive Demonstrations for Robotic Manipulation — arXiv:2608.18787 — arXiv (en), https://arxiv.org/abs/2608.18787 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d4022bc534be (Metraipolis archive dc_d4022bc534be, sha256 4231af103ef61e492c85b138271a485133dd8a672b0f0ef2dd54365bde78a019)
BibTeX
@misc{dcd4022bc534be,
title = {Dream2Reward: Transition-Alignment Reward Models from Positive Demonstrations for Robotic Manipulation — arXiv:2608.18787},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.18787},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d4022bc534be; sha256 4231af103ef61e492c85b138271a485133dd8a672b0f0ef2dd54365bde78a019},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18787. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Learning robotic policies requires dense rewards that remain informative when behavior departs from successful demonstrations. Progress-based rewards estimate how far an observation has advanced along a nominal successful trajectory, but may remain high after an incorrect transition. We introduce Dream2Reward, which learns a language-conditioned successful latent transition field from positive demonstrations. Given the visual history up to a transition start, the model predicts the latent displacement associated with successful execution and scores the observed displacement through signed directional and symmetric magnitude agreement. This transition-level comparison penalizes wrong-direction, overshooting, and stagnant motion even when the resulting observation appears to show progress. Dream2Reward requires no failure annotations, progress labels, or synthetic negatives, and produces a dense causal reward. Across mechanism diagnostics and shared-trajectory evaluations, it provides stronger success-failure separation and more informative feedback on low-quality behavior than progress-based alternatives. Across online and offline policy learning, the same frozen reward model reduces reward hacking and supports stronger downstream performance, including in real-robot manipulation. These results show that comparing realized motion with predicted successful change provides an effective way to convert positive demonstrations into dense rewards for robot learning. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-20 in cs.RO). Categorie arXiv: Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2608.18787 — 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 2026-10-18T14:02:33.850Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).