Pith Number

pith:BDYRVN4B

pith:2026:BDYRVN4BIQ5SUXZXJD33R7UTIW

not attested not anchored not stored refs pending

Controllable LLM Reasoning via Sparse Autoencoder-Based Steering

Boyi Deng, Dayiheng Liu, Fengli Xu, Fuli Feng, Mingfeng Xue, Wenjie Wang, Yi Fang

arxiv:2601.03595 v2 · 2026-01-07 · cs.AI · 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{BDYRVN4BIQ5SUXZXJD33R7UTIW}

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

6 papers in Pith

Discovering Millions of Interpretable Features with Sparse Autoencoders

The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models

Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs

Transcoders Trace Visual Grounding and Hallucinations in Vision-Language Models

Steered Generation via Gradient-Based Optimization on Sparse Query Features

Receipt and verification

First computed	2026-07-29T00:24:34.847515Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

08f11ab781443b2a5f3748f7b8fe93459e3d83bacaf31ea41fb83ee960891575

Aliases

arxiv: 2601.03595 · arxiv_version: 2601.03595v2 · doi: 10.48550/arxiv.2601.03595 · pith_short_12: BDYRVN4BIQ5S · pith_short_16: BDYRVN4BIQ5SUXZX · pith_short_8: BDYRVN4B

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/BDYRVN4BIQ5SUXZXJD33R7UTIW \
  | 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: 08f11ab781443b2a5f3748f7b8fe93459e3d83bacaf31ea41fb83ee960891575

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "67c0d8df4fd3ab1eccf7f0564b400be4acd80e5425ba55e001e2fbc7c984de32",
    "cross_cats_sorted": [
      "cs.CL"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-01-07T05:26:26Z",
    "title_canon_sha256": "c9eff405471f1e6ad3a975d837e5bdf74a4194a193fca89bcf1d44b541f77a1f"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2601.03595",
    "kind": "arxiv",
    "version": 2
  }
}