Stanza: Remote Code Execution via Unsafe Pickle Deserialization in Model Loaders

Stanza is a Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages. Prior to 1.12.2, Stanza model loaders such as stanza.models.common.pretrain.Pretrain.load() attempt torch.load(..., weights_only=True) but fall back to torch.load(..., weights_only=False) on attacker-controllable pickle.UnpicklingError, allowing a malicious .pt pretrain or model file to execute arbitrary pickle code when a Stanza NLP pipeline loads it. This issue is fixed in version 1.12.2.

weblogin/login.fcgi (aka the WebLogin login script) in Stanford University WebAuth 3.5.5, 3.6.0, and 3.6.1 places passwords in URLs in certain circumstances involving conversion of a POST request to a GET request, which allows context-dependent attackers to discover passwords by reading (1) web-server access logs, (2) web-server Referer logs, or (3) the browser history.

Improper Restriction of XML External Entity Reference in stanfordnlp/corenlp

corenlp is vulnerable to Improper Restriction of XML External Entity Reference

Improper Restriction of XML External Entity Reference in stanfordnlp/corenlp

corenlp is vulnerable to Improper Restriction of XML External Entity Reference

stanford-parser v3.9.2 and below was discovered to contain a code injection vulnerability in the component edu.stanford.nlp.io.getBZip2PipedInputStream. This vulnerability is exploited via passing an unchecked argument.

Improper Restriction of XML External Entity Reference in stanfordnlp/corenlp

corenlp is vulnerable to Improper Restriction of XML External Entity Reference

An Incorrect Access Control vulnerability exists in CoreNLP 4.3.2 via the classifier in NERServlet.java (lines 158 and 159).

Improper Restriction of XML External Entity Reference in stanfordnlp/corenlp

corenlp is vulnerable to Improper Restriction of XML External Entity Reference

webauth before 4.6.1 has authentication credential disclosure