CVE-2026-81722: NLTK: Quadratic-time DoS in PorterStemmer via long runs of 'y'
nltk.stem.PorterStemmer.stem() -- a ubiquitous public API applied to arbitrary, often untrusted, tokens -- runs in O(n^2) time on a token containing a long run of the letter 'y', letting a single ~20-50 KB token pin a CPU core (CWE-407). Root cause isconsonant(word, i) was made iterative (commit for #3633, GHSA/CWE-674) to fix an earlier unbounded-recursion RecursionError on 'y'10000.
Published Sep 2, 2026
Affected packages
nltk
- Ecosystem
- PyPI
- Affected
- <= 3.10.2
- Fixed in
- 3.10.3
- Check
pip show nltk
Advisory details
From the GitHub Advisory Database (CC BY 4.0).
nltk.stem.PorterStemmer.stem() -- a ubiquitous public API applied to arbitrary, often untrusted, tokens -- runs in O(n^2) time on a token containing a long run of the letter 'y', letting a single ~20-50 KB token pin a CPU core (CWE-407).
Root cause
_is_consonant(word, i) was made *iterative* (commit for #3633, GHSA/CWE-674) to fix an earlier unbounded-recursion RecursionError on 'y'*10000. The iterative form walks *backward* over the whole run of 'y's on every call:
while i > 0 and word[i] == 'y':
negate = not negate
i -= 1_measure() then calls _is_consonant(stem, i) once for every position i of the stem. For a run of n 'y's that is sum_{i} O(i) = O(n^2). The recursion fix therefore traded a CWE-674 RecursionError for a CWE-407 quadratic-time DoS.
Proof of concept
Measured (Python 3.13): stem('y'*5000 + 'ness') = 2.6s, stem('y'*10000 + 'ness') = 11.3s (2x input -> ~4.3x time = quadratic), stem('y'*20000 + 'ness') > 20s. A pure run of 'y' with no matching suffix is fast because the stemmer rules that call _measure do not fire; a real suffix such as 'ness' triggers _measure on the long stem.
from nltk.stem import PorterStemmer
PorterStemmer().stem('y' * 20000 + 'ness') # >20s of CPUImpact
Stemming is routinely applied to untrusted text (search, indexing, NLP pipelines). A single unbroken ~20-50 KB token of 'y' characters (no whitespace, so it survives tokenization) causes multi-second-to-minutes CPU consumption per request. No confidentiality/integrity impact; single-process availability only.
Fix direction
Classify each character's consonant/vowel status in a single left-to-right O(n) pass (memoise the 'y' run parity) instead of re-walking the run on every _is_consonant call, so _measure and stemming are linear. This is a sibling of the corpus-reader quadratic advisories GHSA-vp2x-qp44-57v7 and GHSA-8mpw-7fpc-4gqj (CWE-407).
References
- https://github.com/nltk/nltk/security/advisories/GHSA-ww6m-cw3f-q94g
- https://nvd.nist.gov/vuln/detail/CVE-2026-81722
- https://github.com/nltk/nltk/commit/7808692d451b962711005d954859bb83aabcf8fa
- https://github.com/nltk/nltk/releases/tag/v3.10.3
- https://github.com/pypa/advisory-database/tree/main/vulns/nltk/PYSEC-2026-3738.yaml
- https://www.vulncheck.com/advisories/nltk-porterstemmer-before-3.10.3-quadratic-time-dos
- https://github.com/advisories/GHSA-ww6m-cw3f-q94g
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
More PyPI vulnerabilities
All PyPI entriesCVE-2026-128135
NLTK: ZIP package extraction follows in-root hardlinks and overwrites files outside the NLTK data root
high7.1Nov 19, 2026CVE-2026-128136
NLTK: AllowlistUnpickler does not deny numpy submodule-local file-I/O sinks
medium6.1Nov 19, 2026CVE-2026-128137
NLTK corpus-reader view methods bypass the scoped root sandbox via intermediate-directory symlinks
medium5.5Nov 19, 2026CVE-2026-128138
NLTK incomplete fix: additional JVM/JAR sandbox bypass through unvalidated launcher tokens in java() command arguments
critical9.8Nov 19, 2026CVE-2026-128139
NLTK: XMLCorpusView comment and CDATA parsing permits quadratic CPU denial of service
high7.5Nov 19, 2026CVE-2026-128140
NLTK: Downloader cyclic collection indices cause an infinite closure walk
medium6.5Nov 19, 2026
