Pith Number

pith:4OFVLZZL

pith:2026:4OFVLZZLOVAQPHVNJWKHMIRTAT

not attested not anchored not stored refs resolved

MLPs are Efficient Distilled Generative Recommenders

Clark Mingxuan Ju, Julian McAuley, Neil Shah, Yupeng Hou, Zitian Guo

Distilling generative recommenders into MLPs preserves accuracy while speeding up inference by 8.74x

arxiv:2605.12617 v1 · 2026-05-12 · 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{4OFVLZZLOVAQPHVNJWKHMIRTAT}

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 SID-MLP matches the accuracy of teacher models while accelerating inference by 8.74x. This distillation strategy can serve as a plug-and-play accelerator for different backbones and tokenizer settings.

C2weakest assumption

The hierarchical nature of SIDs makes prediction difficulty drop sharply after the first token, rendering repeated attention computations highly redundant.

C3one line summary

SID-MLP distills autoregressive generative recommenders into efficient position-specific MLP heads for Semantic ID tasks, achieving 8.74x faster inference with matching accuracy.

References

77 extracted · 77 resolved · 5 Pith anchors

[1] Transformer memory as a differentiable search index 2022

[2] Tran, Jonah Samost, Maciej Kula, Ed H 2023

[3] Adapting large language models by integrating collaborative semantics for recommenda- tion 2024

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

[5] Session-based recommendations with recurrent neural networks 2016

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T03:10:00.495604Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

e38b55e72b7541079ead4d9476223304c9ee1416cd7911867544650c80d1d435

Aliases

arxiv: 2605.12617 · arxiv_version: 2605.12617v1 · doi: 10.48550/arxiv.2605.12617 · pith_short_12: 4OFVLZZLOVAQ · pith_short_16: 4OFVLZZLOVAQPHVN · pith_short_8: 4OFVLZZL

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/4OFVLZZLOVAQPHVNJWKHMIRTAT \
  | 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: e38b55e72b7541079ead4d9476223304c9ee1416cd7911867544650c80d1d435

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "90d565358fe1850a6c46e9f614103872ce80bb47f4c6d8f3850473d21558d259",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.IR",
    "submitted_at": "2026-05-12T18:05:55Z",
    "title_canon_sha256": "805391eadae9bcf536e9256d7da1afd222e74a5616b7d54437935f8d2427f49c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12617",
    "kind": "arxiv",
    "version": 1
  }
}