Recent progress on nuclear parton distribution functions

We report current status of global analyses on nuclear parton distribution functions (NPDFs). The optimum NPDFs are determined by analyzing high‐energy nuclear reaction data. Due to limited experimental measurements, antiquark modifications have large uncertainties at $x > 0.2$ and gluon modifications cannot be determined. A nuclear modification difference between u and d quark distributions could be an origin of the long‐standing NuTeV $sin2θw$ anomaly. There is also an issue of nuclear modification differences between the structure functions of charged‐lepton and neutrino reactions. Next, nuclear clustering effects are discussed in structure functions $F2A$ as a possible explanation for an anomalous result in the $9Be$ nucleus at the Thomas Jefferson National Accelerator Facility (JLab). Last, tensor‐polarized quark and antiquark distribution functions are extracted from HERMES data on the polarized structure function $b1$ of the deuteron, and they could be used for testing theoretical models and for proposing future experiments, for example, the one at JLab. Such measurements could open a new field of spin physics in spin‐one hadrons.