The mayo clinic suggests it could be 70%, and I have seen others suggest 80%. I have not seen anyone suggest as low as 10/20% so I think that suggestion is a huge outlier and even the paper you link says as much.
One of the papers hypothesized that this demonstrates why areas like NYC got hit so hard initially and have remained stable since. Seroprevalence surveys back in the early summer indicated that between 20-30% of NYC residents had contracted Covid already. The lockdowns obviously blunted the surge, but unlike other areas of the country, as NYC has opened back up, there hasn't been much of a second wave at all. With more than 2/3rds of the population still without any kind of exposure to the virus, you would expect a dense city like NYC to see a large increase. The paper concluded that this is evidence that might suggest that the herd immunity barrier is far lower than initially estimated. I've seen some papers suggest something closer to 20-50%, but nothing so low as 10%. I'll see if I can dig up some of the preprints to post here.