Pith Number

pith:NIRAS6TE

pith:2024:NIRAS6TEQZUTBMLGGGHXQESDYQ

not attested not anchored not stored refs pending

Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space

David Dobre, Gauthier Gidel, Leo Schwinn, Sophie Xhonneux, Stephan Gunnemann

arxiv:2402.09063 v2 · 2024-02-14 · 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{NIRAS6TEQZUTBMLGGGHXQESDYQ}

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

3 papers in Pith

LLM-Safety Evaluations Lack Robustness

Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data

Latent Instruction Representation Alignment: defending against jailbreaks, backdoors and undesired knowledge in LLMs

Receipt and verification

First computed	2026-07-05T10:49:43.256997Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

6a22097a64866930b166318f781243c403d443e33b63d6a25221aae88f2f9d33

Aliases

arxiv: 2402.09063 · arxiv_version: 2402.09063v2 · doi: 10.48550/arxiv.2402.09063 · pith_short_12: NIRAS6TEQZUT · pith_short_16: NIRAS6TEQZUTBMLG · pith_short_8: NIRAS6TE

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/NIRAS6TEQZUTBMLGGGHXQESDYQ \
  | 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: 6a22097a64866930b166318f781243c403d443e33b63d6a25221aae88f2f9d33

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f58781ba1a823e5e41e357ca4ac928a8bd348a56533046e56523bfba8edd28f9",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2024-02-14T10:20:03Z",
    "title_canon_sha256": "1f63913efaa848bc296d596831641f53070ec49531fcb2982fca5555e6d35267"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2402.09063",
    "kind": "arxiv",
    "version": 2
  }
}