The authors held scenes constant while changing the instruction through valid target substitutions, arbitrary nouns and unrelated sentences.
Robot policies encode changed instructions but still follow scene cues
Tests in fixed scenes show that alternative valid instructions alter internal representations yet rarely redirect robots toward the newly requested object.
Chinese Tech
Shaohan Jiang · Jiahang Cao · Qiduo He · Fengting Deng · Kun Wu · Jingkai Sun · +6 more
HKU · CUHK(SZ) · Horizon Robotics · Shanghai AI Lab · CUHK
Research Digest··2 min read
Jiang and colleagues tested whether language-conditioned robot policies actually use instructions to choose among multiple feasible actions.
Why this paper
From Shanghai AI Lab and 5 others
In one line
Vision-language robot policies often encode changed instructions but still follow scene-associated targets because language is weakly grounded in action selection.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§