OESA-2026-4045 sentencepiece security update

Security Fixes: Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure.CVE-2026-1260...

OESA-2026-3917 sentencepiece security update

Security Fixes: Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure.CVE-2026-1260...

OESA-2026-3916 sentencepiece security update

Security Fixes: Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure.CVE-2026-1260...

OESA-2026-3915 sentencepiece security update

Security Fixes: Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure.CVE-2026-1260...

sentencepiece:model_load_fuzzer: Crash in sentencepiece::ModelInterface::IsByte

Project: https://github.com/google/sentencepiece.git Detailed Report: https://oss-fuzz.com/testcase?key=6716733566812160 Project: sentencepiece Fuzzing Engine: afl Fuzz Target: modelloadfuzzer Job Type: aflasansentencepiece Platform Id: linux Crash Type: UNKNOWN READ Crash Address: 0x0020bfff8046...

Sentencepiece has a a heap overflow issue

Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure...

PYSEC-2026-1909 Sentencepiece has a a heap overflow issue

Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure...

Fedora 44 : sentencepiece (2026-314504fd18)

The remote Fedora 44 host has a package installed that is affected by a vulnerability as referenced in the FEDORA-2026-314504fd18 advisory. Update to 0.2.1 Tenable has extracted the preceding description block directly from the Fedora security advisory. Note that Nessus has not tested for this...

Fedora 45 : sentencepiece (2026-8755a17c6e)

The remote Fedora 45 host has a package installed that is affected by a vulnerability as referenced in the FEDORA-2026-8755a17c6e advisory. Automatic update for sentencepiece-0.2.1-1.fc45. Changelog Tue May 19 2026 Peter Robinson - 0.2.1-1 - Update to 0.2.1 - Fixes CVE-2026-1260 rhbz2432139...

sentencepiece:trainer_fuzzer: Segv on unknown address in sentencepiece::unigram::Lattice::Viterbi

Project: https://github.com/google/sentencepiece.git Detailed Report: https://oss-fuzz.com/testcase?key=5966429886414848 Project: sentencepiece Fuzzing Engine: afl Fuzz Target: trainerfuzzer Job Type: aflasansentencepiece Platform Id: linux Crash Type: Segv on unknown address Crash Address: Crash...

Security Bulletin: Security vulnerability in Python affects IBM Robotic Process Automation and IBM Robotic Process Automation for Cloud Pak

Summary A security vulnerability in Python affects IBM Robotic Process Automation and IBM Robotic Process Automation for Cloud Pak. Python is used by IBM Robotic Process Automation and IBM Robotic Process Automation for Cloud Pak as part of its deployment. This bulletin identifies the fixes...

sentencepiece:trainer_fuzzer: Segv on unknown address in sentencepiece::unigram::Lattice::Viterbi

Project: https://github.com/google/sentencepiece.git Detailed Report: https://oss-fuzz.com/testcase?key=5298026431447040 Project: sentencepiece Fuzzing Engine: afl Fuzz Target: trainerfuzzer Job Type: aflasansentencepiece Platform Id: linux Crash Type: Segv on unknown address Crash Address: Crash...

OSV-2026-646 Heap-buffer-overflow in sentencepiece::unigram::Model::EncodeOptimized

OSS-Fuzz report: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=507169860 Crash type: Heap-buffer-overflow READ 8 Crash state: sentencepiece::unigram::Model::EncodeOptimized sentencepiece::unigram::Model::Encode sentencepiece::SentencePieceProcessor::Encode...

PT-2026-39310

OSS-Fuzz report: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=507169860 Crash type: Heap-buffer-overflow READ 8 Crash state: sentencepiece::unigram::Model::EncodeOptimized sentencepiece::unigram::Model::Encode sentencepiece::SentencePieceProcessor::Encode...

sentencepiece:model_load_fuzzer: Heap-buffer-overflow in sentencepiece::unigram::Model::EncodeOptimized

Project: https://github.com/google/sentencepiece.git Detailed Report: https://oss-fuzz.com/testcase?key=4785445505662976 Project: sentencepiece Fuzzing Engine: libFuzzer Fuzz Target: modelloadfuzzer Job Type: libfuzzerasansentencepiece Platform Id: linux Crash Type: Heap-buffer-overflow READ 8...

Security Bulletin: Maximo AI Service uses multiple third party dependencies which is vulnerable to multiple CVEs.

Summary Maximo AI Service uses wheel-0.41.3-py3-none-any.whl, orjson-3.10.14-cp311-cp311-manylinux217x8664.manylinux2014x8664.whl, pythonmultipart-0.0.21-py3-none-any.whl, pyasn1-0.6.1.tar.gz, sentencepiece-0.2.0-cp311-cp311-manylinux217x8664.manylinux2014x8664.whl, tar-7.4.3.tgz, tar-7.5.2.tgz...

Google SentencePiece Buffer Overflow Vulnerability

Google SentencePiece is an unsupervised text splitter for neural network-based text generation from Google USA. Google SentencePiece suffers from a buffer overflow vulnerability that stems from an invalid memory access when using a vulnerable model file created by an unusual training process. No...

CVE-2026-1260

A flaw was found in Sentencepiece. This flaw involves invalid memory access when the software processes a specially crafted, vulnerable model file. A local attacker could exploit this by tricking a user into loading such a file. Successful exploitation could lead to a denial of service, informati...

Sentencepiece has a a heap overflow issue

Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure...

GHSA-38VQ-G6VR-W8WF Sentencepiece has a a heap overflow issue

Invalid memory access in Sentencepiece versions less than 0.2.1 when using a vulnerable model file, which is not created in the normal training procedure...