Pith Number

pith:ENNDSNET

pith:2025:ENNDSNET4RK3PYNUS6LTY3KJMR

not attested not anchored not stored refs pending

Using Mechanistic Interpretability to Craft Adversarial Attacks against Large Language Models

Boussad Addad, Katarzyna Kapusta, Thomas Winninger

arxiv:2503.06269 v2 · 2025-03-08 · cs.LG · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{ENNDSNET4RK3PYNUS6LTY3KJMR}

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.

Cited by

5 papers in Pith

Investigating The Security of Modern AI and Cloud Infrastructure

How Do LLMs Cite? A Mechanistic Interpretation of Attribution in Retrieval-Augmented Generation

Subspace-Aware Sparse Autoencoders for Effective Mechanistic Interpretability

Activation-Guided Local Editing for Jailbreaking Attacks

Why Do Large Language Models Generate Harmful Content?

Receipt and verification

First computed	2026-07-05T10:59:06.807402Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

235a393493e455b7e1b497973c6d496455e16fca0d3b13676cf5f84502e74dde

Aliases

arxiv: 2503.06269 · arxiv_version: 2503.06269v2 · doi: 10.48550/arxiv.2503.06269 · pith_short_12: ENNDSNET4RK3 · pith_short_16: ENNDSNET4RK3PYNU · pith_short_8: ENNDSNET

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/ENNDSNET4RK3PYNUS6LTY3KJMR \
  | 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: 235a393493e455b7e1b497973c6d496455e16fca0d3b13676cf5f84502e74dde

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "927c5e58856edf08afb7b7a238f412718c0192f9c56a023d9433c1ffc725824f",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2025-03-08T16:29:45Z",
    "title_canon_sha256": "17198aea37d3de4d9afab909bdab860c77d34dc53010e5296c3d298be823a46e"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2503.06269",
    "kind": "arxiv",
    "version": 2
  }
}