Handling Malformed JSON generated by the Microsoft Academic Knowledge API - json

Question

Making the following request to the Microsoft Academic gives malformed JSON:
https://westus.api.cognitive.microsoft.com/academic/v1.0/interpret?
query=albert einstein&complete=0&count=2&offset=2&model=latest
The result is as follows
Request-Context: appId=cid-v1:7cd06f0e-6d73-46a5-9c10-4ebe681d0156
apim-request-id: 8eec6818-55d3-4b7c-b1c3-b4c6236f0173
Strict-Transport-Security: max-age=31536000; includeSubDomains; preload
x-content-type-options: nosniff
Date: Tue, 31 Oct 2017 01:02:11 GMT
X-Powered-By: ASP.NET
Content-Length: 681
Content-Type: application/json
{
"query": "albert einstein",
"interpretations": [,
{
"logprob": -69.749,
"parse": "<rule name=\"#GetPapers\"><attr name=\"academic#W\">albert</attr> <attr name=\"academic#F.FN\">einstein</attr></rule>",
"rules": [{
"name": "#GetPapers",
"output": {
"type": "query",
"value": "And(W='albert',Composite(F.FN=='einstein'))"
}
}]
},
{
"logprob": -118.575,
"parse": "<rule name=\"#GetPapers\"><attr name=\"academic#W\">albert</attr> <attr name=\"academic#W\">einstein</attr></rule>",
"rules": [{
"name": "#GetPapers",
"output": {
"type": "query",
"value": "And(W='albert',W='einstein')"
}
}]
}
]
}
Notice the leading comma on the third line of the JSON, at the beginning of the interpretations array. I suspect there's a bug in the code used to paginate results. I'm working in Python, and that leading comma breaks every parser I can think of including a straight eval. I think I can handle it with a String.replace before using json.loads, but that's an ugly solution. Are there better alternatives? And am I correct to think this is a bug?