The authors propose GroundingPI, a 4B parameter grounding foundation model designed for broad and precise perception.
Grounding foundation model surpasses larger vision-language models on precise perception and robotics
GroundingPI achieves 73.68% average across 34 grounding benchmarks and improves robot manipulation success rates by up to 24.8% relative to strongest baselines.
Independent
Qize Yu · Lianrui Fan · Boyu Chen · Jiaqi Liang · Xini Ding · Yue Chen · +20 more
Research Digest··3 min read
The authors introduce GroundingPI, a 4-billion-parameter grounding foundation model that generates points and bounding boxes as quantized coordinates in a shared vocabulary.
Why this paper
Independent
In one line
GroundingPI, a 4B model, achieves state-of-the-art visual grounding across 34 benchmarks and improves robotic manipulation and autonomous driving when used as a perception backbone.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks (4 benchmarks)
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§