Pith Number

pith:UFXHD4HB

pith:2026:UFXHD4HBH6SNB3LUIE7MYNLNTD

not attested not anchored not stored refs resolved

Efficient Generative Retrieval for E-commerce Search with Semantic Cluster IDs and Expert-Guided RL

Bokang Wang, Guangxin Song, Jianbo Zhu, Jing Wang, Junjie Bai, Mingmin Jin, Xing Fang, Zhenyu Xie

Category-and-query constrained semantic IDs enable generative retrieval as a practical recall supplement in e-commerce search by halving beam search size and lifting click hit rates.

arxiv:2605.14434 v1 · 2026-05-14 · 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{UFXHD4HBH6SNB3LUIE7MYNLNTD}

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

CQ-SID achieves up to 26.76% and 11.11% relative gains in semantic and personalized click hitrate over RQ-VAE baselines, while halving beam search size. EG-GRPO further improves multi-objective performance. Online A/B tests confirm gains in GMV (+1.15%) and UCTCVR (+0.40%). The generative recall channel now contributes substantially in production, accounting for over 50.25% of exposures, 58.96% of clicks, and 72.63% of purchases.

C2weakest assumption

That the hierarchical semantic cluster IDs produced by category-and-query constrained contrastive learning plus Residual Quantized VAEs preserve sufficient relevance signals to support effective beam search and downstream ranking alignment under the sparse-reward RL regime described.

C3one line summary

CQ-SID semantic IDs and EG-GRPO RL improve generative retrieval hit rates up to 26.76% over RQ-VAE baselines and deliver +1.15% GMV in live e-commerce A/B tests.

References

26 extracted · 26 resolved · 5 Pith anchors

[1] Michele Bevilacqua, Giuseppe Ottaviano, Patrick Lewis, Scott Yih, Sebastian Riedel, and Fabio Petroni. 2022. Autoregressive search engines: Generating substrings as document identifiers.Advances in Ne 2022

[2] arXiv preprint arXiv:2509.03236 , year= 2025

[3] Jiehan Cheng, Zhicheng Dou, Yutao Zhu, and Xiaoxi Li. 2025. Descriptive and discriminative document identifiers for generative retrieval. InProceedings of the AAAI Conference on Artificial Intelligenc 2025

[4] Yingjun Dai and Ahmed El-Roby. 2025. RQ-Rec: Residual Quantized Hierarchical Preference Modeling for Cross-Domain Recommendation. InProceedings of the 33rd ACM International Conference on Multimedia. 2025

[5] OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment 2025 · arXiv:2502.18965

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-17T23:39:07.097082Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

a16e71f0e13fa4d0ed74413ecc356d98ee92d2d3438853165af46a4321ead46e

Aliases

arxiv: 2605.14434 · arxiv_version: 2605.14434v1 · doi: 10.48550/arxiv.2605.14434 · pith_short_12: UFXHD4HBH6SN · pith_short_16: UFXHD4HBH6SNB3LU · pith_short_8: UFXHD4HB

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/UFXHD4HBH6SNB3LUIE7MYNLNTD \
  | 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: a16e71f0e13fa4d0ed74413ecc356d98ee92d2d3438853165af46a4321ead46e

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "3c8188886fb77ba37a1c969c50887122c397947bb9f59318a84cfc8a5ef91053",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-05-14T06:27:46Z",
    "title_canon_sha256": "8626fc65b8ab07af7b888edf88145bd6b7ebe3f6d874179a20e61110cdadb773"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14434",
    "kind": "arxiv",
    "version": 1
  }
}