{"id":"ghsa-ghsa-48m4-24cq-37qf","seq":2056923,"kind":"advisory","title":"ONNX: Out-of-Bounds Read in RNN/GRU/LSTM Shape Inference (Old Opsets, Missing Rank Check)","url":"https://github.com/advisories/GHSA-48m4-24cq-37qf","permalink":"https://vulnpulse.app/items/ghsa-ghsa-48m4-24cq-37qf","summary":"影響: pip:onnx <= 1.21.0 → 1.22.0\nONNX shape inference performed an out-of-bounds read when processing a malformed RNN, GRU, or LSTM node from opsets 1–6 whose first input had a rank other than three. RNNShapeInference_opset1_to_6() checked whether the input shape was present but accessed dim(0) and dim(1) without first checking its rank. A crafted model could therefore terminate a process that called onnx.shape_inference.infer_shapes(). The model passed onnx.checker.check_model() before reaching the vulnerable path. Technical…","severity":"medium","techs":["python"],"cves":["CVE-2026-115919"],"tags":["GHSA-48m4-24cq-37qf","onnx"],"source":{"id":"ghsa","name":"GitHub Advisory Database"},"published_at":"2026-10-28T21:16:24.000Z","fetched_at":"2026-10-29T00:48:52.653Z"}