CVE-2025-3044: LlamaIndex vulnerability in ArxivReader class can cause MD5 hash collisions
A vulnerability in the ArxivReader class of the run-llama/llamaindex repository allows for MD5 hash collisions when generating filenames for downloaded papers. This can lead to data loss as papers with identical titles but different contents may overwrite each other, preventing some papers from being processed for AI model training.
Published Jul 7, 2025 · Updated Jul 8, 2025
Affected packages
llama-index-readers-papers
- Ecosystem
- PyPI
- Affected
- < 0.3.1
- Fixed in
- 0.3.1
- Check
pip show llama-index-readers-papers
Advisory details
From the GitHub Advisory Database (CC BY 4.0).
A vulnerability in the ArxivReader class of the run-llama/llama_index repository allows for MD5 hash collisions when generating filenames for downloaded papers. This can lead to data loss as papers with identical titles but different contents may overwrite each other, preventing some papers from being processed for AI model training. The issue is resolved in llama-index-readers-papers version 0.3.1 (in llama-index 0.12.28).
References
- https://nvd.nist.gov/vuln/detail/CVE-2025-3044
- https://github.com/run-llama/llama_index/commit/0008041e8dde8e519621388e5d6f558bde6ef42e
- https://huntr.com/bounties/80182c3a-876f-422f-8bac-38267e0345d6
- https://github.com/run-llama/llama_index/commit/f69e1c0e7579228fec4cfaf716e4f951e131de77
- https://github.com/advisories/GHSA-p7j4-jwjf-5x9w
How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology
