Any idea why Alibaba might ship a model while aware that it has this defect when it is so easy to fix?

Is this looping tough to fix by changing the training?

Model tuning is like a black box.

To put it simply, it's like pressing the first button and the second one pops up, pressing the second button and the third one pops up, then pressing the third button and both the first and second ones pop up.

Nobody knows why, but it just happens.

And when doing post-training, problems like overfitting, catastrophic forgetting, output loops, and so on will all appear.

To achieve balanced results, you can only aim for general performance rather than bringing out the model's full capabilities.

Also, it's very possible that they really didn't notice these problems, because model evaluation is really difficult, teams and resources are limited, and many finer details are only discovered later.

Steering/model weight lobotomy usually have a ton of side effects. You fix one defect and add 10 others.