CVE-2025-71011
Unknown Unknown - Not Provided
BaseFortify

Publication date: 2026-01-29

Last updated on: 2026-02-03

Assigner: MITRE

Description
An input validation vulnerability in the flow.Tensor.new_empty/flow.Tensor.new_ones/flow.Tensor.new_zeros component of OneFlow v0.9.0 allows attackers to cause a Denial of Service (DoS) via a crafted input.
CVSS Scores
EPSS Scores
Probability:
Percentile:
Meta Information
Published
2026-01-29
Last Modified
2026-02-03
Generated
2026-05-07
AI Q&A
2026-01-29
EPSS Evaluated
2026-05-05
NVD
EUVD
Affected Vendors & Products
Showing 1 associated CPE
Vendor Product Version / Range
oneflow oneflow 0.9.0
Helpful Resources
Exploitability
CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-20 The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required to process the data safely and correctly.
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?

This vulnerability is an input validation flaw in the OneFlow deep learning framework version 0.9.0, specifically in the tensor creation functions flow.Tensor.new_empty(), flow.Tensor.new_ones(), and flow.Tensor.new_zeros(). When these functions are called with negative tensor dimensions, which are invalid, the internal logic calculating the total number of elements (elem_cnt) results in a negative value. This causes the program to fail internal checks and abort, leading to a crash. For example, flow.Tensor.new_empty() accepts a negative dimension initially but crashes later, while flow.Tensor.new_ones() and flow.Tensor.new_zeros() immediately abort with a core dump. This results in a denial-of-service condition by causing the OneFlow process to terminate unexpectedly. [1]


How can this vulnerability impact me? :

This vulnerability can cause a denial-of-service (DoS) condition in applications using the affected OneFlow tensor creation APIs. By providing crafted inputs with negative tensor dimensions, an attacker can cause the OneFlow process to abort unexpectedly, leading to crashes and instability. This can disrupt the availability and reliability of services or applications relying on OneFlow for deep learning tasks. [1]


How can this vulnerability be detected on my network or system? Can you suggest some commands?

This vulnerability can be detected by testing the OneFlow tensor creation functions with negative dimensions to observe if the process crashes or aborts. For example, running Python code that calls flow.Tensor.new_empty(), flow.Tensor.new_ones(), or flow.Tensor.new_zeros() with negative dimensions (e.g., -2) will trigger the issue. A sample test command in Python could be: ```python import oneflow as flow # This should cause a crash or abort if vulnerable flow.Tensor.new_ones([-2]) ``` Monitoring for core dumps or process aborts when such inputs are used can help detect the vulnerability. [1]


What immediate steps should I take to mitigate this vulnerability?

Immediate mitigation steps include validating all inputs to the tensor creation functions to ensure that no negative dimensions are passed. Avoid using untrusted or malformed inputs that could contain negative dimensions. Additionally, consider updating OneFlow to a version where this vulnerability is fixed once available. Until then, implement input validation checks in your application code to prevent negative dimension values from reaching the vulnerable functions. [1]


Ask Our AI Assistant
Need more information? Ask your question to get an AI reply (Powered by our expertise)
0/70
EPSS Chart