Pith Number

pith:Q32VHWDU

pith:2017:Q32VHWDUNWUVJG4NGXMOSJ6K2M

not attested not anchored not stored refs resolved

BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain

Brendan Dolan-Gavitt, Siddharth Garg, Tianyu Gu

An adversary can train a neural network that performs well on normal inputs but activates malicious behavior on specific attacker-chosen triggers.

arxiv:1708.06733 v2 · 2017-08-22 · cs.CR · cs.LG

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{Q32VHWDUNWUVJG4NGXMOSJ6K2M}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

an adversary can create a maliciously trained network (a backdoored neural network, or a BadNet) that has state-of-the-art performance on the user's training and validation samples, but behaves badly on specific attacker-chosen inputs.

C2weakest assumption

The attacker must have sufficient control over the training process or data to embed the backdoor without detection, as assumed in the outsourced training scenario described.

C3one line summary

Adversaries can create backdoored neural networks during outsourced training that maintain high accuracy on normal data but misbehave on attacker-chosen triggers.

References

53 extracted · 53 resolved · 2 Pith anchors

[1] ImageNet large scale visual recognition competition 2012

[2] Speech recognition with deep recurrent neural networks 2013

[3] Multilingual Distributed Representations without Word Alignment 2014 · arXiv:1312.6173

[4] Neural machine translation by jointly learning to align and translate 2014

[5] Playing atari with deep reinforce- ment learning 2013

Formal links

2 machine-checked theorem links

Cited by

90 papers in Pith

The Power of Backdoor Absorption in Community Training

Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks

TooBad: Backdoor Diffusion Models with Ultra-Low Poison Rate and Imperceptible Trigger

Rethinking Molecular Graph Backdoors under Chemistry-aware Admission

CLIP-guided Diffusion Model for Backdoor Generation in Sensor-based Human Activity Recognition

Receipt and verification

First computed	2026-07-04T23:27:34.718401Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

86f553d8746da9549b8d35d8e927cad32a1b38e24e428e9d64899184a0e3c15a

Aliases

arxiv: 1708.06733 · arxiv_version: 1708.06733v2 · doi: 10.48550/arxiv.1708.06733 · pith_short_12: Q32VHWDUNWUV · pith_short_16: Q32VHWDUNWUVJG4N · pith_short_8: Q32VHWDU

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/Q32VHWDUNWUVJG4NGXMOSJ6K2M \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 86f553d8746da9549b8d35d8e927cad32a1b38e24e428e9d64899184a0e3c15a

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "43366791bf313c93af4a47b23d2f9edc58e1862c9386c98a090808dcd58ef384",
    "cross_cats_sorted": [
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CR",
    "submitted_at": "2017-08-22T17:31:54Z",
    "title_canon_sha256": "2d6d46e9e11448172b302ae907d89d20653060d9233ae659a7ec1537a22532fb"
  },
  "schema_version": "1.0",
  "source": {
    "id": "1708.06733",
    "kind": "arxiv",
    "version": 2
  }
}