PulseAugur
EN
LIVE 09:49:47

Kuaishou deploys new retrieval model to boost video watch time

By PulseAugur Editorial · [1 sources] ·

Researchers have developed a Controllable Retrieval Model (CRM) designed to enhance efficiency in recommendation systems by incorporating desired watch times as a condition. This framework, implemented using both deep neural networks and a Decision Transformer, allows for steering retrieval without retraining by inputting a target watch time. CRM has been deployed in Kuaishou's short-video recommendation system, leading to measurable improvements in user watch time. AI

IMPACT This new retrieval model could enhance user engagement in short-video platforms by personalizing content recommendations based on desired watch times.

RANK_REASON Publication of a research paper detailing a new AI model and its application. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Kuaishou deploys new retrieval model to boost video watch time

How we ranked this

Signal score
12 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Publication of a research paper detailing a new AI model and its application. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.AI TIER_1 English(EN) · Chi Liu, Jiangxia Cao, Rui Huang, Kuo Cai, Weifeng Ding, Qiang Luo, Kun Gai, Guorui Zhou ·

    CRM: Retrieval Model with Controllable Condition

    arXiv:2412.13844v2 Announce Type: replace-cross Abstract: Retrieval models for recommendation need to account for outcomes such as watch time while searching large item corpora efficiently. We present the Controllable Retrieval Model (CRM), an outcome-conditioned retrieval framew…

RELATED ENTITIES

RELATED TOPICS