Metraipolis › archive › dc_ea5e0cc96f3c

GIFT: Goal-Injected Fine-Tuning for Efficient Manipulation Policy Adaptation — arXiv:2609.07006

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07006
License (as archived)
CC0-1.0
Archived
2026-10-16T15:11:33.846Z
SHA-256 of the archived text
9dd623d9884d4bd0a6a1e273356f3938ba05fbb1dcf20548548a93ab5b306712
Stable URL of this page
https://metraipolis.com/documento/dc_ea5e0cc96f3c

How to cite

GIFT: Goal-Injected Fine-Tuning for Efficient Manipulation Policy Adaptation — arXiv:2609.07006 — arXiv (en), https://arxiv.org/abs/2609.07006 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea5e0cc96f3c (Metraipolis archive dc_ea5e0cc96f3c, sha256 9dd623d9884d4bd0a6a1e273356f3938ba05fbb1dcf20548548a93ab5b306712)
BibTeX
@misc{dcea5e0cc96f3c,
  title = {GIFT: Goal-Injected Fine-Tuning for Efficient Manipulation Policy Adaptation — arXiv:2609.07006},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07006},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea5e0cc96f3c; sha256 9dd623d9884d4bd0a6a1e273356f3938ba05fbb1dcf20548548a93ab5b306712},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07006. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Compared with relying solely on initial observations and language instructions, predicting goal images with generative models as high-level visual guidance can significantly enhance the robustness of Vision-Language-Action (VLA) models. However, most existing foundation models have not systematically incorporated goal image conditioning due to the high computational training cost. To this end, we propose Goal-Injected Fine-Tuning (GIFT), a lightweight and efficient fine-tuning framework that seamlessly integrates generated goal images into multiple representative pretrained VLA models. Our approach introduces goal image features into observations via a zero-initialized convolution which progressively grows parameters from zero and prevents harmful noise from disrupting the pretrained policy during fine-tuning. As training proceeds, goal information is gradually incorporated, enabling efficient goal understanding without disrupting model stability. We further introduce a refined image editing method to generate semantically and visually consistent goal images from initial observations and task instructions. Experiments show that goal-aware VLA models achieve substantial performance gains across tasks: with only a single epoch of fine-tuning, GIFT outperforms the base model by 6.0% and 13.4% on two SIMPLER settings, and by 4.7% on LIBERO, demonstrating both efficiency and effectiveness.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.07006 — 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 2026-10-16T14:49:57.164Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).