SoK: Attack and Defense Landscape of Mobile On-device AI Systems | Paper Citation Record

Observation fbf1f660-4587-4b83-a6cd-b7ffba46bad7 · outbound

This paper cites https://developer.apple.com/documentation/avfoundation/avc am-building-a-camera-app, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developer.apple.com/documentation/avfoundation/avc am-building-a-camera-app, 2026

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.468230Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 1bcc91ea-d686-45aa-9b12-763db3f76af7 · outbound

This paper cites developer.android.com/media/camera/camerax, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems developer.android.com/media/camera/camerax, 2026

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.493597Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8ef08c7b-0afd-4da8-a8b4-87de52a5efe6 · outbound

This paper cites developer.apple.com/machine-learning/core-ml/, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems developer.apple.com/machine-learning/core-ml/, 2026

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.409047Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 25d43070-6c24-4ecb-a407-f470e11dbc4d · outbound

This paper cites https://executorch.ai/, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://executorch.ai/, 2026

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.382365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 186958bd-2e58-43d6-8ba7-e92f1ac23dd1 · outbound

This paper cites https://developers.google.com/edge/litert-lm/mode ls/gemma-4, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developers.google.com/edge/litert-lm/mode ls/gemma-4, 2026

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.501025Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0108ec1d-843a-45b0-b2aa-f2872d3f9ce1 · outbound

This paper cites https://developers.google.com/edge/litert/next/tenso r-sdk, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developers.google.com/edge/litert/next/tenso r-sdk, 2026

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.491900Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 09be9827-0f92-4cf2-91b6-397ca90cf3ce · outbound

This paper cites https://ai.google.dev/edge/litert, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://ai.google.dev/edge/litert, 2026

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.503138Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation fedc2f78-2d64-4bca-9878-525c015919c8 · outbound

This paper cites https://developer.apple.com/documentation/coreml/m lfeaturevalue, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developer.apple.com/documentation/coreml/m lfeaturevalue, 2026

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.467995Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 93a454e5-8d2b-4be1-bd7b-0cce4bb982dd · outbound

This paper cites https://apple.fandom.com/wiki/Neural Engine, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://apple.fandom.com/wiki/Neural Engine, 2026

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.448956Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation bbbe2bdc-d4b8-45ed-92d5-f449477ba504 · outbound

This paper cites https://developers.googleblog.com/on-device-genai -in-chrome-chromebook-plus-and-pixel-watch-with-litert-lm/, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developers.googleblog.com/on-device-genai -in-chrome-chromebook-plus-and-pixel-watch-with-litert-lm/, 2026

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.499176Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation ca911525-2da9-4a1c-8810-60dfa22a90b0 · outbound

This paper cites https://developers.google.com/edge/lit ert/conversion/tensorflow/build/ondevice training, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://developers.google.com/edge/lit ert/conversion/tensorflow/build/ondevice training, 2026

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.495252Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 13a266c6-3a62-4941-8dcd-2a06c8c65ea8 · outbound

This paper cites https://onnx.ai/, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://onnx.ai/, 2026

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.465721Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 66a54461-2fa1-47f0-9208-ddb934b7ef94 · outbound

This paper cites www.qualcomm.com/processors/hexagon, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems www.qualcomm.com/processors/hexagon, 2026

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.490325Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c8598d6e-40b5-4469-943f-9e8055d4a960 · outbound

This paper cites https://ai.google.dev/edge/api/tflite/java/org/tensorflow/ lite/support/image/TensorImage, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems https://ai.google.dev/edge/api/tflite/java/org/tensorflow/ lite/support/image/TensorImage, 2026

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.528914Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3ab93a04-3f54-4c21-b969-91c2296263a2 · outbound

This paper cites Offline model guard: secure and private ml on mobile devices.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Offline model guard: secure and private ml on mobile devices

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.532785Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 92b5f735-1b09-49aa-8e29-b39b5e2ae46e · outbound

This paper cites Efficient compositional multi-tasking for on-device large language models.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Efficient compositional multi-tasking for on-device large language models

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.508927Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 125f479b-d70b-4ac4-aeb4-97a086717232 · outbound

This paper cites Sanctuary: Arming trustzone with user-space enclaves.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Sanctuary: Arming trustzone with user-space enclaves

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.446186Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 66f0fd2f-34b9-4859-bf80-2b84dc1de464 · outbound

This paper cites Improving end-to-end neural diarization using conversational summary representations.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Improving end-to-end neural diarization using conversational summary representations

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.461164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b8a66f6f-bcb5-4354-b47c-d7acec847c94 · outbound

This paper cites Cheating your apps: Black-box adversarial attacks on deep learning apps.Journal of Software: Evolution and Process, 36(4):e2528, 2024.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Cheating your apps: Black-box adversarial attacks on deep learning apps.Journal of Software: Evolution and Process, 36(4):e2528, 2024

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.448019Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9c5e6edf-ae9f-460e-821c-b4b3f543a1d8 · outbound

This paper cites Guardiann: Fast and secure on- device inference in trustzone using embedded sram and cryptographic hardware.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Guardiann: Fast and secure on- device inference in trustzone using embedded sram and cryptographic hardware

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.497327Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6a056406-88b6-457c-a922-f15b52feae11 · outbound

This paper cites Understanding real-world threats to deep learning models in android apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Understanding real-world threats to deep learning models in android apps

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.453543Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation d5f3cf13-4987-445b-b92c-6daaaaa68a33 · outbound

This paper cites Hybridtee: Secure mobile dnn execution using hybrid trusted execution environment.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Hybridtee: Secure mobile dnn execution using hybrid trusted execution environment

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.453959Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 4d8df976-224a-45b3-a4f2-b211edfeeaa0 · outbound

This paper cites Secure and efficient mobile dnn using trusted execution environments.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Secure and efficient mobile dnn using trusted execution environments

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.530912Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9f7f768f-8ee4-4ffb-a4c5-57dbad6d75c1 · outbound

This paper cites A first look at on-device models in ios apps.ACM Transactions on Software Engineering and Methodology, 33(1):1–30, 2023.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems A first look at on-device models in ios apps.ACM Transactions on Software Engineering and Methodology, 33(1):1–30, 2023

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.482906Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation de5f3506-4589-434e-a77e-a7853fb38f37 · outbound

This paper cites Mmguard: Automatically protecting on-device deep learning models in android apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Mmguard: Automatically protecting on-device deep learning models in android apps

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.492060Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 4842fd4c-c409-41ef-a1da-796bc8d804ed · outbound

This paper cites Malmodel: Hiding malicious payload in mo- bile deep learning models with black-box backdoor attack.Automated Software Engineering, 33(1):28, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Malmodel: Hiding malicious payload in mo- bile deep learning models with black-box backdoor attack.Automated Software Engineering, 33(1):28, 2026

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.499012Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9ecea03d-7bb3-4d97-beeb-e073a9867341 · outbound

This paper cites Smart app attack: hacking deep learning models in android apps.IEEE Transactions on Information Forensics and Security, 17:1827–1840, 2022.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Smart app attack: hacking deep learning models in android apps.IEEE Transactions on Information Forensics and Security, 17:1827–1840, 2022

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.437338Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation eb755615-860e-4379-b90b-ce5081d0d0e7 · outbound

This paper cites Robustness of on- device models: Adversarial attack to deep learning models on android apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Robustness of on- device models: Adversarial attack to deep learning models on android apps

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.439185Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5b2242bd-1797-4f92-bd88-e99470dd023c · outbound

This paper cites Typhon unleashed: Practical adversarial weight attacks against on-device deep learning models.IEEE Transactions on Dependable and Secure Computing, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Typhon unleashed: Practical adversarial weight attacks against on-device deep learning models.IEEE Transactions on Dependable and Secure Computing, 2026

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.470441Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 65f4bfb1-d2e1-4270-9190-a45f871053a4 · outbound

This paper cites Themis: Towards practical intellectual property protection for post-deployment on-device deep learning models.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Themis: Towards practical intellectual property protection for post-deployment on-device deep learning models

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.381078Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 68b473df-f4e2-471b-af51-24378e1c4e92 · outbound

This paper cites TinyML Security: Exploring Vulnerabilities in Resource-Constrained Machine Learning Systems.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems TinyML Security: Exploring Vulnerabilities in Resource-Constrained Machine Learning Systems

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:46:54.731712Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 45313ff8-851a-4373-82c6-908386eb432a · outbound

This paper cites Confidential execution of deep learning inference at the untrusted edge with arm trustzone.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Confidential execution of deep learning inference at the untrusted edge with arm trustzone

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.416425Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 2790b0cd-f662-41cc-8a67-2173817b120d · outbound

This paper cites an unresolved cited work.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Unresolved cited work

Reference 33

Resolution: unresolved; raw_fallback, observed 2026-07-06T02:51:52.431443Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f9a3dcaa-56ec-4805-abcc-6f55b3862051 · outbound

This paper cites Redlc: Learning-driven reverse engi- neering for deep learning compilers.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Redlc: Learning-driven reverse engi- neering for deep learning compilers

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.452122Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 05596338-5a9d-45a6-81b9-a4dffddefe2d · outbound

This paper cites Efficient layout- guided image inpainting for mobile use.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Efficient layout- guided image inpainting for mobile use

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.463914Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f6687c49-a4b5-4e78-aa34-0dd605476cf3 · outbound

This paper cites Deeppayload: Black-box backdoor attack on deep learning models through neural payload injection.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Deeppayload: Black-box backdoor attack on deep learning models through neural payload injection

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.500803Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6e00d371-4c85-4230-8eb0-e2af2ec042f9 · outbound

This paper cites Model extraction attack against on-device deep learning with power side channel.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Model extraction attack against on-device deep learning with power side channel

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.537604Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9e267bbd-f2fe-4118-8ebc-857296854dcd · outbound

This paper cites Secdeep: Secure and performant on-device deep learning inference framework for mobile and iot devices.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Secdeep: Secure and performant on-device deep learning inference framework for mobile and iot devices

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.474122Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3408ef87-11b8-47d1-b8e8-884a56f8cfee · outbound

This paper cites Deepcache: Revisiting cache side-channel attacks in deep neural networks executables.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Deepcache: Revisiting cache side-channel attacks in deep neural networks executables

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.484787Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a5393100-adfd-4bde-a16a-f5eea5a4add5 · outbound

This paper cites Mir- rornet: A tee-friendly framework for secure on-device dnn inference.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Mir- rornet: A tee-friendly framework for secure on-device dnn inference

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.488576Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation bac75a12-9436-4f5c-b42b-771b32498afd · outbound

This paper cites Quantization backdoors to deep learning commercial frame- works.IEEE Transactions on Dependable and Secure Computing, 21(3):1155–1172, 2023.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Quantization backdoors to deep learning commercial frame- works.IEEE Transactions on Dependable and Secure Computing, 21(3):1155–1172, 2023

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.411027Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 46c223d5-0ead-4b35-9106-3adaa5e371d7 · outbound

This paper cites Darknetz: towards model privacy at the edge using trusted execution environments.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Darknetz: towards model privacy at the edge using trusted execution environments

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.486497Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b59ce579-b7d4-4a0e-bc4a-ea14444a3825 · outbound

This paper cites A novel obfuscation method based on majority logic for preventing unauthorized access to binary deep neural networks.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems A novel obfuscation method based on majority logic for preventing unauthorized access to binary deep neural networks

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.535109Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3d632b2e-4b6e-4922-a1ec-5acce60547c4 · outbound

This paper cites Asgard: Protecting on-device deep neural networks with virtualization-based trusted execution environments.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Asgard: Protecting on-device deep neural networks with virtualization-based trusted execution environments

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.404505Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 52d4bb52-e079-4f17-b3e7-6d936bf935eb · outbound

This paper cites In33rd USENIX Security Symposium (USENIX Security 24), pages 5233–5250, 2024.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems In33rd USENIX Security Symposium (USENIX Security 24), pages 5233–5250, 2024

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.476412Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 91321b04-3c90-4de2-9c24-0e2a82d81873 · outbound

This paper cites Demistify: Identifying on- device machine learning models stealing and reuse vulnerabilities in mobile apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Demistify: Identifying on- device machine learning models stealing and reuse vulnerabilities in mobile apps

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.463135Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 03678be1-38fe-44bf-89b1-1b4e87068819 · outbound

This paper cites Beyond the model: Data pre-processing attack to deep learning models in android apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Beyond the model: Data pre-processing attack to deep learning models in android apps

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.493760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 055bb628-f7ea-46cb-9b7e-c1651351fc87 · outbound

This paper cites Leap: Trustzone based developer-friendly tee for intelligent mobile apps.IEEE Transactions on Mobile Computing, 22(12):7138–7155, 2022.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Leap: Trustzone based developer-friendly tee for intelligent mobile apps.IEEE Transactions on Mobile Computing, 22(12):7138–7155, 2022

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.526524Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b7376a79-6ad9-4356-a4aa-6decd02d1131 · outbound

This paper cites Tensorshield: Safeguarding on-device inference by shielding critical dnn tensors with tee.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Tensorshield: Safeguarding on-device inference by shielding critical dnn tensors with tee

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.482517Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8924e178-e934-4442-aaec-870fcfb5100a · outbound

This paper cites Tsqp: Safeguarding real-time inference for quantization neural networks on edge devices.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Tsqp: Safeguarding real-time inference for quantization neural networks on edge devices

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.490164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8573dc09-6941-4226-9013-b9097047d362 · outbound

This paper cites Shadownet: A secure and efficient on-device model inference system for convolutional neural networks.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Shadownet: A secure and efficient on-device model inference system for convolutional neural networks

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.402401Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 89f6a1f3-ea98-4e56-8f2c-e6ab2b087e00 · outbound

This paper cites Mind your weight (s): A large-scale study on insufficient machine learning model protection in mobile apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Mind your weight (s): A large-scale study on insufficient machine learning model protection in mobile apps

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.458901Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0afee317-d47a-4fd6-a7eb-e47d3405e701 · outbound

This paper cites Game of arrows: On the ({In-) Security}of weight obfuscation for{On-Device}{TEE- Shielded}{LLM}partition algorithms.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Game of arrows: On the ({In-) Security}of weight obfuscation for{On-Device}{TEE- Shielded}{LLM}partition algorithms

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.415225Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9b359fc9-c706-4439-9ff8-f1decc48e5ba · outbound

This paper cites Tz-llm: Protecting on-device large language models with arm trustzone.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Tz-llm: Protecting on-device large language models with arm trustzone

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.451291Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8720f3b8-436b-439e-8d17-b7ebac0cd760 · outbound

This paper cites Energy- latency attacks to on-device neural networks via sponge poisoning.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Energy- latency attacks to on-device neural networks via sponge poisoning

Reference 55

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.507103Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 4e99cfdb-600b-4cbb-ad54-b7e694a6789e · outbound

This paper cites Stealthy Backdoor Attack to Real-world Models in Android Apps.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Stealthy Backdoor Attack to Real-world Models in Android Apps

Reference 56

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:46:54.740085Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 4091fec3-23fd-4c94-b3b5-89812c2d3089 · outbound

This paper cites SoK: Towards Security and Safety of Edge AI.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems SoK: Towards Security and Safety of Edge AI

Reference 57

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:46:54.732160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c9a36fdb-0449-4647-9b96-f2738a63cb46 · outbound

This paper cites Tim: Enabling large-scale white- box testing on in-app deep learning models.IEEE Transactions on Information Forensics and Security, 19:8188–8203, 2024.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Tim: Enabling large-scale white- box testing on in-app deep learning models.IEEE Transactions on Information Forensics and Security, 19:8188–8203, 2024

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.505204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 24ae92dc-f483-42e7-b833-653c679c6927 · outbound

This paper cites FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation

Reference 59

Resolution: verified exact; local_arxiv, observed 2026-07-02T11:46:54.734607Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 4f135295-bff4-435b-a1f1-02c1e8627754 · outbound

This paper cites A first look at deep learning apps on smartphones.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems A first look at deep learning apps on smartphones

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.497054Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c163e46f-310e-4f80-b45c-64e684844620 · outbound

This paper cites Groupcover: A secure, efficient and scalable inference framework for on-device model protection based on tees.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Groupcover: A secure, efficient and scalable inference framework for on-device model protection based on tees

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.470026Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 7b42aa7a-cde4-422d-aff3-6621aaa975ba · outbound

This paper cites No privacy left outside: On the (in-) security of tee-shielded dnn partition for on-device ml.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems No privacy left outside: On the (in-) security of tee-shielded dnn partition for on-device ml

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.458216Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a609b582-b9d6-4042-96b0-abd5fac9a816 · outbound

This paper cites Miragenet: A secure, efficient, and scalable on-device model protection in heterogeneous tee and gpu system.arXiv preprint arXiv:2601.13826, 2026.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Miragenet: A secure, efficient, and scalable on-device model protection in heterogeneous tee and gpu system.arXiv preprint arXiv:2601.13826, 2026

Reference 63

Resolution: verified exact; arxiv_id, observed 2026-07-02T11:46:54.737344Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 9f0c7e26-bbaa-4a2a-9f66-ffee9436c9d2 · outbound

This paper cites Dynamo: Protecting mobile dl models through coupling obfuscated dl operators.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Dynamo: Protecting mobile dl models through coupling obfuscated dl operators

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.472474Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation d5635e1e-c166-4c25-8f41-9c03f3f629b8 · outbound

This paper cites Model-less is the best model: Generating pure code implementations to replace on-device dl models.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Model-less is the best model: Generating pure code implementations to replace on-device dl models

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.443680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 3ec70fa4-8fb1-45ec-b88f-14ff5b1cbdbf · outbound

This paper cites Modelobfuscator: Obfuscating model information to protect deployed ml-based systems.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Modelobfuscator: Obfuscating model information to protect deployed ml-based systems

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.495548Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a40bea54-c209-48c5-83f9-1a21eb312165 · outbound

This paper cites Investigating white-box attacks for on-device models.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Investigating white-box attacks for on-device models

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.441239Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation ed4f6995-52da-407e-a092-47c0bdef3c18 · outbound

This paper cites Nnsplitter: an active defense solution for dnn model via automated weight obfuscation.

SoK: Attack and Defense Landscape of Mobile On-device AI Systems Nnsplitter: an active defense solution for dnn model via automated weight obfuscation

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-07-06T02:51:52.433040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers