Pith. sign in

REVIEW 5 cited by

LONGER: Scaling Up Long Sequence Modeling in Industrial Recommenders

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2505.04421 v2 pith:TODNBCSM submitted 2025-05-07 cs.IR

classification cs.IR
keywords longermodelingservingattentionbytedancefullyindustriallong
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Modeling ultra-long user behavior sequences is critical for capturing both long- and short-term preferences in industrial recommender systems. Existing solutions typically rely on two-stage retrieval or indirect modeling paradigms, incuring upstream-downstream inconsistency and computational inefficiency. In this paper, we present LONGER, a Long-sequence Optimized traNsformer for GPU-Efficient Recommenders. LONGER incorporates (i) a global token mechanism for stabilizing attention over long contexts, (ii) a token merge module with lightweight InnerTransformers and hybrid attention strategy to reduce quadratic complexity, and (iii) a series of engineering optimizations, including training with mixed-precision and activation recomputation, KV cache serving, and the fully synchronous model training and serving framework for unified GPU-based dense and sparse parameter updates. LONGER consistently outperforms strong baselines in both offline metrics and online A/B testing in both advertising and e-commerce services at ByteDance, validating its consistent effectiveness and industrial-level scaling laws. Currently, LONGER has been fully deployed at more than 10 influential scenarios at ByteDance, serving billion users.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. UniRank: Benchmarking Ranking Models for Unified Sequential Modeling and Feature Interaction

    cs.IR 2026-07 conditional novelty 6.0 of 10

    UniRank is an open benchmark that standardizes chronological autoregressive supervision, multi-task evaluation, and capacity controls for 15 unified ranking models on five large datasets.

  2. Towards Comprehensible Recommendation with Large Language Model Fine-tuning

    cs.IR 2025-08 conditional novelty 6.0 of 10

    CURec aligns an LLM with recommendation objectives via RL, generates personalized user patterns and item-reason texts, and corrects them through a chronological loop, improving top-K recall and NDCG on MovieLens and Amazon.

  3. RankMixer: Scaling Up Ranking Models in Industrial Recommenders

    cs.IR 2025-07 conditional novelty 6.0 of 10

    RankMixer scales an industrial ranking model to 1B dense parameters with 10x MFU improvement and unchanged latency, gaining 1.08% in app duration in Douyin A/B tests.

  4. Beyond Quality: Unlocking Diversity in Ad Headline Generation with Large Language Models

    cs.CL 2025-08 conditional novelty 5.0 of 10

    DIVER trains an LLM with synthetic data plus multi-objective RL to emit several diverse, style-tagged ad headlines in one pass, reporting online ADVV and CTR gains.

  5. Large Foundation Model for Ads Recommendation

    cs.LG 2025-08 conditional novelty 5.0 of 10

    Tencent's LFM4Ads transfers user, item, and user-item cross representations from a pre-trained foundation model into downstream ad models via feature, module, and model-level mechanisms, reporting a 2.45% platform-wid...

Pith tools