Pith. sign in

REVIEW 15 cited by

QARM: Quantitative Alignment Multi-Modal Recommendation at Kuaishou

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2411.11739 v1 pith:VBVAAIU7 submitted 2024-11-18 cs.IR cs.AI

Xinchen Luo , Jiangxia Cao , Tianyu Sun , Jinkai Yu , Rui Huang , Wei Yuan , Hezheng Lin , Yichen Zheng
show 12 more authors
Shiyao Wang Qigen Hu Changqing Qiu Jiaqi Zhang Xu Zhang Zhiheng Yan Jingming Zhang Simin Zhang Mingxing Wen Zhaojie Liu Kun Gai Guorui Zhou
This is my paper
classification cs.IRcs.AI
keywords multi-modalmodeldownstreamrecommendationmodelsrepresentationrepresentationstasks
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In recent years, with the significant evolution of multi-modal large models, many recommender researchers realized the potential of multi-modal information for user interest modeling. In industry, a wide-used modeling architecture is a cascading paradigm: (1) first pre-training a multi-modal model to provide omnipotent representations for downstream services; (2) The downstream recommendation model takes the multi-modal representation as additional input to fit real user-item behaviours. Although such paradigm achieves remarkable improvements, however, there still exist two problems that limit model performance: (1) Representation Unmatching: The pre-trained multi-modal model is always supervised by the classic NLP/CV tasks, while the recommendation models are supervised by real user-item interaction. As a result, the two fundamentally different tasks' goals were relatively separate, and there was a lack of consistent objective on their representations; (2) Representation Unlearning: The generated multi-modal representations are always stored in cache store and serve as extra fixed input of recommendation model, thus could not be updated by recommendation model gradient, further unfriendly for downstream training. Inspired by the two difficulties challenges in downstream tasks usage, we introduce a quantitative multi-modal framework to customize the specialized and trainable multi-modal information for different downstream models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 15 Pith papers

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

  1. Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning

    cs.AI 2026-05 unverdicted novelty 6.0 of 10

    Search-E1 uses GRPO interleaved with on-policy self-distillation to reach 0.440 average EM on seven QA benchmarks with Qwen2.5-3B, outperforming open-source baselines.

  2. FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation

    cs.AI 2026-05 unverdicted novelty 6.0 of 10

    FLUID introduces LUCID semantic codes from a multimodal encoder to retire item IDs in livestreaming rankers, with staged warmup yielding online gains of +0.55% watch duration and +2.05% cold-start views.

  3. Bi-Level Optimization for Generative Recommendation: Bridging Tokenization and Generation

    cs.IR 2025-10 unverdicted novelty 6.0 of 10

    BLOGER is a bi-level optimization framework that jointly optimizes the tokenizer and recommender for generative recommendation, outperforming prior methods on real-world datasets.

  4. Understanding Generative Recommendation with Semantic IDs from a Model-scaling View

    cs.AI 2025-09 conditional novelty 6.0 of 10

    Semantic-ID-based generative recommenders saturate as model size grows, while directly using an LLM as the recommender keeps improving with scale and learns collaborative filtering signals better.

  5. Empowering Large Language Model for Sequential Recommendation via Multimodal Embeddings and Semantic IDs

    cs.IR 2025-09 conditional novelty 6.0 of 10

    MME-SID improves LLM-based sequential recommendation by fusing collaborative, text, and image embeddings with quantized semantic IDs, using MMD reconstruction and code-embedding initialization.

  6. Generative Recommendation with Semantic IDs: A Practitioner's Handbook

    cs.IR 2025-07 conditional novelty 6.0 of 10

    An open-source framework and ablation study showing which semantic-ID generative recommendation components actually matter, with results that challenge several standard defaults.

  7. VRAgent-R1: Boosting Video Recommendation with MLLM-based Agents via Reinforcement Learning

    cs.MM 2025-07 conditional novelty 6.0 of 10

    VRAgent-R1 uses an MLLM agent to summarize videos and a reinforcement-learned agent to simulate user choices, improving video recommendation and user-decision simulation on MicroLens-100K.

  8. Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning

    cs.AI 2026-05 unverdicted novelty 5.0 of 10

    Search-E1 interleaves vanilla GRPO with offline self-distillation via token-level forward KL alignment to privileged sibling trajectories, reaching 0.440 average EM on seven QA benchmarks with Qwen2.5-3B and beating o...

  9. FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation

    cs.AI 2026-05 unverdicted novelty 5.0 of 10

    FLUID retires candidate-side item IDs in production livestream rankers via cross-domain multimodal hierarchical codes and late-fusion ID-free design, reporting online gains of +0.55% Quality Watch Duration and +2.05% ...

  10. ReST: A Plug-and-Play Spatially-Constrained Representation Enhancement Framework for Local-Life Recommendation

    cs.IR 2025-11 unverdicted novelty 5.0 of 10

    ReST enhances long-tail item representations for spatially constrained local-life recommendations via a Meta ID Warm-up Network and a contrastive SIDENet with hard sampling and dynamic alignment strategies.

  11. Diffusion-based Multi-modal Synergy Interest Network for Click-through Rate Prediction

    cs.IR 2025-08 conditional novelty 5.0 of 10

    Diff-MSIN improves multi-modal CTR prediction AUC by roughly 1.5-2% over prior multi-modal baselines by disentangling common, specific, and synergistic modality features.

  12. MISS: Multi-Modal Tree Indexing and Searching with Lifelong Sequential Behavior for Retrieval Recommendation

    cs.IR 2025-08 conditional novelty 5.0 of 10

    MISS builds a k-means index tree on interaction-supervised multi-modal embeddings and adds two behavior search units (Co-GSU, MM-GSU) plus ESU/MMoE, reporting ~30-47% relative recall gains over TDM+MMoE on Kuaishou da...

  13. OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment

    cs.IR 2025-02 unverdicted novelty 5.0 of 10

    OneRec unifies retrieval and ranking in a generative recommender using session-wise decoding and iterative DPO-based preference alignment, achieving real-world gains on Kuaishou.

  14. Dynamic Content Moderation in Livestreams: Combining Supervised Classification with MLLM-Boosted Similarity Matching

    cs.CV 2025-12 conditional novelty 4.0 of 10

    A deployed hybrid moderation system combining supervised classification and reference-based similarity matching, boosted by MLLM distillation, reduces unwanted livestream views by 6–8%.

  15. GR-LLMs: Recent Advances in Generative Recommendation Based on Large Language Models

    cs.IR 2025-07 unverdicted novelty 3.0 of 10

    A survey of LLM-based generative recommendation systems, covering application settings, training pipelines, industrial deployment challenges, and future directions.

Pith tools