Pith Number

pith:HXICGPKP

pith:2026:HXICGPKPZ5QHLYIDZNDQKUNONG

not attested not anchored not stored refs resolved

Rich-Media Re-Ranker: A User Satisfaction-Driven LLM Re-ranking Framework for Rich-Media Search

Feicheng Li, Jianxin Li, Ligang Zhou, Qingyun Sun, Ying Nie, Zeyang Tang, Zhiming Peng, Zihao Guo

The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度.

arxiv:2602.05408 v2 · 2026-02-05 · cs.IR

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{HXICGPKPZ5QHLYIDZNDQKUNONG}

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

Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines. Notably, the proposed framework has been deployed in a large-scale industrial search system, yielding substantial improvements in online user engagement rates and satisfaction metrics.

C2weakest assumption

That the VLM-based evaluator and LLM-based re-ranker, after multi-task reinforcement learning, produce reliable holistic scores on relevance, quality, novelty, information gain, and visual presentation without systematic bias or overfitting to the training distribution.

C3one line summary

A re-ranking system for rich-media search that plans query intents from sessions, adds visual signals from VLMs, and uses an LLM to score results on multiple facets before multi-task RL adaptation, with reported gains in engagement after industrial deployment.

References

50 extracted · 50 resolved · 8 Pith anchors

[1] Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, and Adam Jatowt

[2] Dear: Dual-stage document reranking with reasoning agents via llm distillation 2025

[3] Qingyao Ai, Keping Bi, Jiafeng Guo, and W Bruce Croft. 2018. Learning a deep listwise context model for ranking refinement. InThe 41st international ACM SIGIR conference on research & development in i 2018

[4] Seq2slate: Re-ranking and slate optimization with rnns 2018 · arXiv:1810.02019

[5] Heng-Tze Cheng, Levent Koc, Jeremiah Harmsen, Tal Shaked, Tushar Chandra, Hrishi Aradhye, Glen Anderson, Greg Corrado, Wei Chai, Mustafa Ispir, et al

Receipt and verification

First computed	2026-05-18T02:44:31.591454Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1

Aliases

arxiv: 2602.05408 · arxiv_version: 2602.05408v2 · doi: 10.48550/arxiv.2602.05408 · pith_short_12: HXICGPKPZ5QH · pith_short_16: HXICGPKPZ5QHLYID · pith_short_8: HXICGPKP

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/HXICGPKPZ5QHLYIDZNDQKUNONG \
  | 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: 3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "a7fd24e7f97b21ef428f7a9a29a546f4371968ddec51bf6e373b714004243f5b",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-02-05T07:45:42Z",
    "title_canon_sha256": "c251f592e2868d8fa68c79b7c523b3c4cc30ee8a0f3f57a4623a62fd550686ed"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2602.05408",
    "kind": "arxiv",
    "version": 2
  }
}