Pengchuan Zhang
speaker
270 appearances
1 recordings
1 series
first heard Dec 2025
last heard 18 Dec
Pengchuan Zhang’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsRecordings per month over the last 12 months — 1 in all, peaking in Dec 2025 with 1.
Appearances
Then kinda if kinda the exhaustivity is filled, then we can
to next step.
You can see that we can go to the kind of the pipeline, go to this kind of so-called kind of human manual correction.
Human kind of manually annotate kind of all these kind of missing masks to make this data point exhaustive.
So you can see that
Exhaustivity is a very big factor there and we play it as the kind of center place in this data engine.
And uh
But you consider if we ask human annotator to annotate every mass from scratch, it will take a lot of time.
I remember kinda each data point in the beginning will take about more than kinda two minutes to finish.
But if you use model in the loop, then it's reduced to about kinda five forty-five seconds.
Can I you can use model to propose mass and then just human to kinda to annotate?
Another very key kind of innovation in this data engine is that we really find that these verification steps, like to verify a mask is good or not, or to verify now the good mass are exhaustive or not, can be done by AI, can be done by
that multimodal model.
That is a breakthrough, and then kind of we can fun tune our kind of for example Namas 3.2 with our kind of verification, human annotated verification data.
We get kind of superhuman performance on these two verification tasks.
And then we do not need human on these two tasks.
This further brings our kind of per data point annotation time to about can 25 seconds.
So you can
Is that from the original kind of all human to kind of about two minutes to final 25 minutes for one kind of data point?
How can a this is kind of our gonna
Showing 81–100 of 270 · page 5 of 14
← Previous
Next →