TacInno-VLA fine-tunes π₀.₅ for contact-rich robot manipulation. It uses images, instructions, robot state and touch to generate actions. A tactile tokenizer supplies contact-dynamics targets during training. The first release covers six UniVTAC policies and a four-task Franka FR3 policy.

기본 정보

모델 유형
행동 정책
아키텍처
VLA
행동 생성
플로 매칭
등록된 버전
TacInno-VLA

논문 및 보고서 (1)

TacInno: Teaching Vision-Language-Action Models the Contact Dynamics Beyond Vision and Action

논문 있음

arXiv: 2611.26405
저자 및 소속 (8)

저자 순서: 원문 순서

  1. Weile Liequal_contribution

    ShanghaiTech University · Institute of AI for Industries (IAII), Chinese Academy of Sciences

  2. Yunfan Louequal_contribution

    Peking University

  3. Muqing Shiequal_contribution

    Washington University in St. Louis

  4. Zihao Zhang

    Institute of AI for Industries (IAII), Chinese Academy of Sciences

  5. Zhenbiao Huang

    Beijing FiveAges Technology

  6. Boxian Yao

    Peking University

  7. Xiyu Zhu

    Peking University

  8. Zhibo Pangcorresponding_author

    Peking University

최초 공개
2026-11-20
최신 논문 개정
v1 · 2026-11-20
학회 / 학술지
arXiv

관련 모델 버전

논문 ↗PDF ↗

공식 자료

출처 또는 수정 제안 보내기