Pith Number

pith:6UZGADZM

pith:2025:6UZGADZMYHG6NVBEIAIFLUXH7X

not attested not anchored not stored refs pending

Understanding Refusal in Language Models with Sparse Autoencoders

Clement Neo, Erik Cambria, Nirmalendu Prakash, Ranjan Satapathy, Roy Ka-Wei Lee, Wei Jie Yeo

arxiv:2505.23556 v1 · 2025-05-29 · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{6UZGADZMYHG6NVBEIAIFLUXH7X}

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

4 papers in Pith

Faithfulness to Refusal: A Causal Audit of Neuron Selectors

OpenSafeIntent: Evaluating Intent-Calibrated Safe Completion Across Dual-Use Prompt Sets

Do Activation Monitors Survive Model Updates? Benchmarking, Predicting, and Repairing Activation-Monitor Staleness

Beyond "I cannot fulfill this request": Alleviating Rigid Rejection in LLMs via Label Enhancement

Receipt and verification

First computed	2026-07-05T11:12:03.862380Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

f532600f2cc1cde6d424401055d2e7fddaa2530b4ebbce14d5729e9f4a077a3a

Aliases

arxiv: 2505.23556 · arxiv_version: 2505.23556v1 · doi: 10.48550/arxiv.2505.23556 · pith_short_12: 6UZGADZMYHG6 · pith_short_16: 6UZGADZMYHG6NVBE · pith_short_8: 6UZGADZM

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/6UZGADZMYHG6NVBEIAIFLUXH7X \
  | 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: f532600f2cc1cde6d424401055d2e7fddaa2530b4ebbce14d5729e9f4a077a3a

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f8977356cb7d094d15c1bd682b0c120ac401f76d4709be723c7fc30fadc4acc1",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-05-29T15:33:39Z",
    "title_canon_sha256": "9fb52cffbc396e5ae8968e55ce022d4154bc71c629fdd3e0b3f544dead9e2803"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2505.23556",
    "kind": "arxiv",
    "version": 1
  }
}