Frontend
GTI: Grounding Task Intent Throughout Long-Horizon VLA Manipulation
arXiv cs.RO0 views
arXiv:2611.23926v1 Announce Type: new Abstract: Vision-language-action (VLA) models generate local actions from visual observations, language, and robot state, but long-horizon manipulation requires the global task intent to remain grounded in the currently executable context. We propose GTI (Grounding Task Intent), which uses the current observation and execution history to localize the active subtask and its execution phase. For the same local subtask, GTI generates two phase-dependent visual targets: an interaction/contact state and a local completion state. The local language instruction and the corresponding phase-specific visual target jointly condition the VLA, specifying both what to do now and what state should be reached at the current phase. During execution, anomaly monitoring and local recovery re-establish an intent-consistent execution context through rollback, re-observation, and phase re-evaluation after persistent deviation. We evaluate GTI on RoboCasa365, VLA-Arena Horizon-Long, and real-world PiPER tasks. Relative to matched Global-FT baselines, GR00T success on seen composite RoboCasa365 tasks increases from 14.6%/12.1% to 31.8%/20.4% under ID/OOD evaluation. On VLA-Arena, GTI+$\pi_{0.5}$ reaches 47.6% overall success versus 45.2% for an adapted ForeAct baseline, with gains of 5.2/6.0 percentage points on the harder L1/L2 tasks. In the three real-robot task experiments, mean full-task success increases from 21.7% to 59.2%. The results indicate that persistent semantic intent localization and phase-conditioned visual grounding improve multi-stage execution, while failure-triggered local recovery provides additional deployment-time support.
Source:https://arxiv.org/abs/2611.23926PreviousReLoco: Unified Humanoid Locomotion and Fall Recovery via Behavior-Conditioned Motion PriorsNextA Hierarchical Onboard Vision-Language-Navigation Framework for Aerial Robots in Complex Environments
Related posts
We found a bug in our own ring size chart, so we open-sourced the fix
0dev.to
Keeping a Live Stream and Its Game State in Sync
0dev.to
Evaluating AI-Generated Code: Beyond Syntax and Correctness
0dev.to
Engineering practice: git, code, tests · 3. Code review: what to look for, in what order
0dev.to
Rate Limiting a Real-Time Odds Feed Without Dropping Critical Updates
0dev.to
Localising Payment Flows for Central Asian Markets
0dev.to UX