From your measurement curves, there are three artefacts in your graphs. One in the UV region as you mentioned. Another two are in the regions of near IR and middle IR respectively. These could be l
caused by one or more of several factors below:
1. Lamp changing causes the incident light shining on the sample not on the same positions before the lamp changed and after the lamp changed;
2. The aperatures before the sample and on the reference beam need to be identical or very much similar. This could be corrected by running the baseline.
3. The aperatures could be bigger than your sample; it should be smaller than a sample for the measurements.
4. Check the system to make sure the system calibrated well.