Pith Number

pith:4TKVYUQ3

pith:2026:4TKVYUQ35CIMP6EFZW2EEDAWDD

not attested not anchored not stored refs pending

HeadRank: Decoding-Free Passage Reranking via Preference-Aligned Attention Heads

Aolin Li, Chenxing Wang, Dongliang Liao, Haijun Wu, Huiyun Hu, Jin Xu, Junwu Du, Juyuan Wang, Ligang Liu, Shunlin Rong, Yuchen Fang

HeadRank reranks passages by aligning LLM attention heads to preferences in continuous space without decoding.

arxiv:2604.17237 v2 · 2026-04-19 · cs.IR · 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{4TKVYUQ35CIMP6EFZW2EEDAWDD}

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

HeadRank consistently outperforms generative and decoding-free baselines with 100% formatting success. At 4B, 57.4% of relevant middle-zone documents reach the top quartile versus 14.2% for irrelevant ones -- a 43-percentage-point selectivity gap that demonstrates the effectiveness of attention-space preference alignment for listwise reranking.

C2weakest assumption

That entropy-regularized head selection combined with hard adjacent-level preference pairs and a distribution regularizer can reliably overcome attention homogenization in middle context using only 211 training queries without introducing new biases or overfitting to the small training set.

C3one line summary

HeadRank improves decoding-free passage reranking by preference-aligning attention heads to increase discriminability in middle-context documents, outperforming baselines on 14 benchmarks with only 211 training queries.

Receipt and verification

First computed	2026-05-20T01:05:13.893425Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

e4d55c521be890c7f885cdb4420c1618eac1c35bef825b35821ce0dc4a4c10b3

Aliases

arxiv: 2604.17237 · arxiv_version: 2604.17237v2 · doi: 10.48550/arxiv.2604.17237 · pith_short_12: 4TKVYUQ35CIM · pith_short_16: 4TKVYUQ35CIMP6EF · pith_short_8: 4TKVYUQ3

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/4TKVYUQ35CIMP6EFZW2EEDAWDD \
  | 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: e4d55c521be890c7f885cdb4420c1618eac1c35bef825b35821ce0dc4a4c10b3

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "44ad9b279e33eb1e6c26b522cd6cda643171e9178f61986c6cf3972ea785d4e9",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-04-19T03:43:42Z",
    "title_canon_sha256": "460dfc3d414688b56310e270bb82f31d4742ed23010bf53b87b7a06001d8e8ef"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.17237",
    "kind": "arxiv",
    "version": 2
  }
}