Pith Number

pith:KKFBMZGH

pith:2026:KKFBMZGHARJKFW7WUMB3VIV7G7

not attested not anchored not stored refs resolved

Conditional Attribute Estimation with Autoregressive Sequence Models

Andrew J. Loza, Daniella Meeker, Erica Stutz, Giacomo Marino, Qiao Liu

Conditional Attribute Transformers estimate sequence attributes from each possible next token in one forward pass.

arxiv:2605.14004 v1 · 2026-05-13 · 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{KKFBMZGHARJKFW7WUMB3VIV7G7}

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

Our approach achieves state of the art performance on sparse reward tasks, improves next-token prediction at sufficient model sizes, estimates attribute probabilities orders of magnitude faster than sampling, and can guide decoding of autoregressive sequence models on a range of language tasks.

C2weakest assumption

The framework assumes that sequence-level attributes can be accurately estimated from partial sequences and single next-token conditionals without requiring full-sequence rollouts or additional supervision during training.

C3one line summary

Conditional Attribute Transformers jointly estimate next-token probabilities and conditional attribute values for autoregressive sequence models, enabling credit assignment, counterfactuals, and steerable generation in one pass.

References

37 extracted · 37 resolved · 4 Pith anchors

[1] Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901 1901

[2] Protgpt2 is a deep unsupervised language model for protein design.Nature communications, 13(1):4348 2022

[3] Generative medical event models improve with scale 2025

[4] Genome modeling and design across all domains of life with evo 2.BioRxiv, pages 2025–02, 2025 2025

[5] Training Compute-Optimal Large Language Models 2022 · arXiv:2203.15556

Receipt and verification

First computed	2026-05-17T23:39:13.128457Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

528a1664c70452a2dbf6a303baa2bf37ed5782974cf42ef0cbb72f5223c5e8b0

Aliases

arxiv: 2605.14004 · arxiv_version: 2605.14004v1 · doi: 10.48550/arxiv.2605.14004 · pith_short_12: KKFBMZGHARJK · pith_short_16: KKFBMZGHARJKFW7W · pith_short_8: KKFBMZGH

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/KKFBMZGHARJKFW7WUMB3VIV7G7 \
  | 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: 528a1664c70452a2dbf6a303baa2bf37ed5782974cf42ef0cbb72f5223c5e8b0

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "d00c9fbd3b8badf57911c87242f74136a27e3e703bd91fe3ddf9a5c9e34482c0",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-13T18:11:16Z",
    "title_canon_sha256": "6241970baf562565c6fbe6ff8eb3e7e39a15c82ecca4b3d8e768a6c4604cb163"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14004",
    "kind": "arxiv",
    "version": 1
  }
}