I’m not the OP, and haven’t read much Marcus, but an analysis linking “chains of claims” together could be interesting, devaluing specific true claims when they

I’m not the OP, and haven’t read much Marcus, but an analysis linking “chains of claims” together could be interesting, devaluing specific true claims when they are used to support a false claim. The claims dataset appears to evaluate each claim independently.