Hugging Face peft's LoRA-GA and CorDA initialization modules (src/peft/tuners/lora/corda.py lines ~102 and ~163, and src/peft/tuners/lora/loraga.py line ~101) call torch.load on config-specified cache/covariance files without weights_only=True, bypassing peft's own safe-loading wrapper used elsewhere in the codebase.
{
"cna_assigner": "TuranSec",
"cwe_ids": [
"CWE-502"
],
"osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/71xxx/CVE-2026-71281.json"
}{
"extracted_events": [
{
"range": [
{
"introduced": "0"
},
{
"last_affected": "0.19.1"
}
],
"source": "AFFECTED_FIELD",
"strategy": "StandardRange"
}
]
}