The rules of evidence don't apply at sentencing. FRE 1101(d)(3). But that doesn't mean that anything goes. At the very least, sentencing evidence must be reliable, whether it comes from police officers, codefendants, girlfriends, drug addicts, or confidential informants. See, e.g., United States v. Padilla, 793 Fed. Appx 749 (10th Cir. 2019) (police report); United States v. Dahda, 852 F.3d 1282 (10th Cir. 2017) (codefendants); United States v. Fennell, 65 F.3d 812 (10th Cir. 1995) (girlfriend); United States v. Richards, 27 F.3d 465 (10th Cir. 1994) (drug addict); United States v. Ortiz, 993 F.2d 204 (10th Cir. 1993) (confidential informant); accord USSG 6A1.3(a) ("In resolving any dispute concerning a factor important to the sentencing determination, the court may consider relevant information without regard to its admissibility under the rules of evidence applicable at trial, provided that the information has sufficient indicia of reliability to support its probable accuracy.") (emphasis added).
To that end, why not argue that scientific, technical, and specialized evidence should be subject to something like a Daubert inquiry before your sentencing court relies on it? Think risk-assessment tools, mental-health evaluations (especially those done in jail/prison settings), traditional forensic evidence, cash-equivalency assessments, drug-quantity approximations---you get the picture. After all, reliability is the touchstone of Daubert. Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579, 589 (1993) ("[t]he trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable").
For more on this front, check out Maneka Sinha, Junk Science at Sentencing, 89 Geo. Wash. L. Rev. 52 (2021).
And remember: Object to any statements in the PSR that lack a reliable evidentiary source, and put the government to its burden of proving reliability at sentencing.