Open Source
HINT: Human-Intent Inception for Agentic Robot Manipulation
arXiv cs.RO0 views
arXiv:2609.02653v3 Announce Type: replace Abstract: Humans can perform complex manipulations given a simple intent through an overall instruction, while continuously adapting to evolving visual observations. However, current vision-language action (VLA) models and other action policies struggle to realize this high-level intelligent behavior under dense, evolving visual inputs and sparse language guidance. Visual correlations can then dominate semantic intent, leading actions to follow visual shortcuts rather than human goals. We present HINT (Human-INTent INcepTion), an agentic framework inspired by the human manipulation principles: semantic intent changes sparsely at manipulation-pattern transitions, whereas continuous control primarily depends on the evolving object-hand relationship. HINT invokes semantic reasoning only at pattern transitions to resolve the current subtask and target, then maintains this commitment through multi-view grounding and visual tracking. We explore two visual interfaces-image-space semantic highlighting and attention-prior injection-to communicate the tracked intent to the action policy without introducing additional trainable parameters into the foundation action model. Experiments across three long-horizon tasks and out-of-distribution variants show that HINT substantially improves intent understanding, task progress, and end-to-end success across two foundation policies while preserving low-latency control. Project page: https://robot-hint.github.io/
Source:https://arxiv.org/abs/2609.02653PreviousContact-Aided Factor-Graph Localization for Underwater SamplingNextNavi-Agent: Unlocalized Monocular Navigation Agent
Related posts
How to get a software engineering internship
0dev.to Career

Got a new MacBook for college? 9 apps I'd install before the first assignment (2026)
0dev.to Beginners
Four order books, one clock: what a six-second probe says about Hyperliquid's latency claims
0dev.to OpenSource
App licences: MIT, GPL, proprietary — where projects actually break
0dev.to Programming
regex-engine: a regex engine from scratch in Rust, with no backtracking
0dev.to Rust
MOSU2: Long-range Navigation System with Multi-modal Scene Understanding
0arXiv cs.RO