Regular expression to find files with underscores and optional extension

This is for work, so I've changed the extensions and files to protect the innocent.

I am parsing text from a description looking for a file name in the format word_here and it can have as many underscores as needed plus an optional extension. I was able to come up with this regular expression which works

Test 1

text = 'Some text here: * my_file_stuff.mat * other_file * third_file *'

FILE_REG_EX = r'([\w]+_+[\w]+\.*[py|mat]*)'
res = re.findall(FILE_REG_EX, text)

print(res)

Output 1

python test_regex.py

['my_file_stuff.mat', 'other_file', 'third_file']

The problem is it doesn't work for stuff like this

Test 2

text = '|my_file|another_file.mat|O_HERES_ONE|_O_HERES_ANOTHER| | | |'

FILE_REG_EX = r'([\w]+_+[\w]+\.*[py|mat]*)'
res = re.findall(FILE_REG_EX, text)

print(res)

Output 2

python test_regex.py
['my_file|a', 'nother_file.mat|', 'O_HERES_ONE|', '_O_HERES_ANOTHER|']

I modified my regex to include the vertical bar, here

Test 3

text = '|my_file|another_file.mat|O_HERES_ONE|_O_HERES_ANOTHER| | | |'

FILE_REG_EX = r'([\w]+_+[\w]+\.*[py|plot]*)\|'
res = re.findall(FILE_REG_EX, text)

print(res)

Output 3

 python test_regex.py
['my_file', 'another_file.mat', 'O_HERES_ONE', 'O_HERES_ANOTHER']

and that works for the second one but now not for the first one. Part of the issue is I will be searching a description for text to look up where a file is at, and I have no way of knowing what formatting it will use for files, only that they will be something in the form of MY_FILE_HERE01.py with or without the extension.

I've tried using the not symbol to exclude the vertical bars in front and back, but that seems to come up empty for both strings.