BestHub
Sign in
Tagged articles
105 articles
Page 1 of 2
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Apr 18, 2026 · Artificial Intelligence

Model Ability Gets Squeezed Out in Multi‑Task Learning—How ESM Preserves It (CVPR 2026)

The paper reveals that multi‑task models suffer performance drops because tasks compete for the same internal subspace, and introduces Essential Subspace Merging (ESM) which separates critical directions and uses Polarized Scaling to keep multiple abilities stable, achieving significantly lower degradation than traditional baselines.

ESDESMessential subspace
0 likes · 16 min read
Model Ability Gets Squeezed Out in Multi‑Task Learning—How ESM Preserves It (CVPR 2026)
AIWalker
AIWalker
Mar 22, 2026 · Artificial Intelligence

Can a Single Vision Model Replace Multiple Specialized Networks? Nvidia’s New Aggregated Foundation Model

Nvidia’s latest aggregated vision foundation model consolidates detection, segmentation, and other visual tasks into one network, eliminating the complexity and resource waste of multi‑model stacks; the article explains the challenges of resolution balance and teacher distribution, outlines three model generations (RADIOv2.5, C‑RADIOv3, C‑RADIOv4), and details the novel multi‑teacher distillation techniques that boost performance across benchmarks.

Model AggregationNvidiaknowledge distillation
0 likes · 6 min read
Can a Single Vision Model Replace Multiple Specialized Networks? Nvidia’s New Aggregated Foundation Model
HyperAI Super Neural
HyperAI Super Neural
Mar 12, 2026 · Artificial Intelligence

Stanford’s Merlin: Single‑GPU 3D Abdominal CT Vision‑Language Model Leads 752 Tasks

Stanford researchers introduced Merlin, the first native 3D abdominal CT vision‑language foundation model trained on a single NVIDIA A6000 GPU using a 25,494‑scan dataset, and demonstrated its superiority across 752 benchmark tasks—including zero‑shot classification, phenotype prediction, cross‑modal retrieval, disease forecasting, report generation, and 3D segmentation—outperforming existing baselines.

3D CTDisease PredictionVision-Language Model
0 likes · 18 min read
Stanford’s Merlin: Single‑GPU 3D Abdominal CT Vision‑Language Model Leads 752 Tasks
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Mar 6, 2026 · Artificial Intelligence

Why Reasoning and Tool-Use Clash in Agentic RL—and How DART Solves It

Recent studies reveal that in Agentic RL, jointly training reasoning and tool-use on shared parameters creates a persistent negative interaction, with gradients nearly orthogonal, limiting performance; a disentangled tuning approach (DART) using separate LoRA adapters isolates the two abilities and restores gains across benchmarks.

DARTGradient InterferenceLoRA
0 likes · 12 min read
Why Reasoning and Tool-Use Clash in Agentic RL—and How DART Solves It
AIWalker
AIWalker
May 26, 2025 · Artificial Intelligence

VisionReasoner: RL‑Unified Model Beats YOLO‑World Detection, Segmentation, Counting

VisionReasoner presents a reinforcement‑learning‑driven unified framework that simultaneously tackles detection, segmentation, and counting tasks, employing a novel multi‑target cognition strategy and efficient Hungarian‑based matching, and demonstrates substantial gains—29.1% on COCO detection, 22.1% on ReasonSeg, and 15.3% on CountBench—using only 7,000 training samples.

Reinforcement LearningSegmentationVisionReasoner
0 likes · 20 min read
VisionReasoner: RL‑Unified Model Beats YOLO‑World Detection, Segmentation, Counting
JD Tech
JD Tech
May 6, 2025 · Artificial Intelligence

One4All Generative Recommendation Framework for CPS Advertising

This article reviews recent advances in applying large language models to CPS advertising recommendation, outlines business requirements and core technical challenges, proposes an extensible multi‑task generative framework with explicit intent perception and multi‑objective optimization, and presents offline and online performance gains along with future research directions.

AI OptimizationCPS advertisingGenerative Models
0 likes · 13 min read
One4All Generative Recommendation Framework for CPS Advertising
JD Retail Technology
JD Retail Technology
Apr 2, 2025 · Artificial Intelligence

One4All: A Scalable Multi‑Task Generative Recommendation Framework for CPS Advertising

The paper introduces One4All, a scalable multi‑task generative recommendation framework for CPS advertising that combines few‑shot intent prompting, a Rewards‑in‑Context multi‑objective optimization, and an online model‑selection strategy, delivering 2‑3× offline HitRate/NDCG gains and notable online CTR, CVR, and commission improvements.

AdvertisingLLMLarge Language Models
0 likes · 14 min read
One4All: A Scalable Multi‑Task Generative Recommendation Framework for CPS Advertising
AI Frontier Lectures
AI Frontier Lectures
Mar 21, 2025 · Artificial Intelligence

How ConFIG Eliminates Gradient Conflicts for Faster Multi‑Task Deep Learning

The paper introduces ConFIG (Conflict‑Free Inverse Gradients), a mathematically proven method that resolves gradient conflicts among multiple loss terms in physics‑informed neural networks, multi‑task learning, and continual learning, and its momentum‑based variant M‑ConFIG that further accelerates training while maintaining accuracy.

CONFIGGradient ConflictM-ConFIG
0 likes · 11 min read
How ConFIG Eliminates Gradient Conflicts for Faster Multi‑Task Deep Learning
AntTech
AntTech
Jan 13, 2025 · Artificial Intelligence

Two Ant Group Papers Selected for AAAI 2025: Human‑Feedback Evaluation Framework for Product Image Background Inpainting and Bagging‑Expert Network for Multi‑Task Learning

Two Ant Group papers accepted at AAAI 2025—one presenting a human‑feedback‑driven evaluation framework for product image background inpainting using EfficientSAM and a new HFPC‑44k dataset, and the other proposing a Bagging‑Expert Network to mitigate expert polarization in multi‑gate mixture‑of‑experts for multi‑task learning.

AAAI 2025Ant GroupBagging-Expert Network
0 likes · 4 min read
Two Ant Group Papers Selected for AAAI 2025: Human‑Feedback Evaluation Framework for Product Image Background Inpainting and Bagging‑Expert Network for Multi‑Task Learning
Zhuanzhuan Tech
Zhuanzhuan Tech
Nov 6, 2024 · Artificial Intelligence

Multi-Task Learning for E-commerce Search: Overview, Practices, and Model Design in the Zhuanzhuan Scenario

This article reviews the necessity, benefits, and practical implementations of multi-task learning in e‑commerce search, detailing model selection, architecture extensions such as ESMM and ESM², and future directions for handling user behavior sequences and multi‑objective optimization.

Deep LearningESMMModel architecture
0 likes · 13 min read
Multi-Task Learning for E-commerce Search: Overview, Practices, and Model Design in the Zhuanzhuan Scenario
NewBeeNLP
NewBeeNLP
Oct 21, 2024 · Artificial Intelligence

Why Do MOE Experts Collapse? An In‑Depth Look at HOME’s Multi‑Task Architecture

This article analyzes the polarization issues in industrial Mixture‑of‑Experts (MoE) frameworks, explains expert collapse, degradation, and under‑fitting, and details the HOME model’s input types, architectural innovations, normalization, gating mechanisms, and related DICE‑BN insights.

Expert NormalizationGating MechanismsMixture of Experts
0 likes · 10 min read
Why Do MOE Experts Collapse? An In‑Depth Look at HOME’s Multi‑Task Architecture
Tencent Advertising Technology
Tencent Advertising Technology
Jul 17, 2024 · Artificial Intelligence

Ads Recommendation in a Collapsed and Entangled World: Tencent's Innovations in Feature Encoding, Dimensional Collapse Mitigation, and Interest Disentanglement

This article summarizes Tencent Advertising's recent research on recommendation models, covering comprehensive feature encoding techniques, solutions to embedding dimensional collapse through multi‑embedding paradigms, and novel methods such as STEM and AME to disentangle conflicting user interests across multiple tasks.

AdvertisingEmbeddingdimensional collapse
0 likes · 20 min read
Ads Recommendation in a Collapsed and Entangled World: Tencent's Innovations in Feature Encoding, Dimensional Collapse Mitigation, and Interest Disentanglement
DataFunSummit
DataFunSummit
Apr 15, 2024 · Artificial Intelligence

Deep Learning Practices for Internet Real‑Estate Recommendation at 58.com

This article details the end‑to‑end deep‑learning pipeline used by 58.com for real‑estate recommendation, covering business background, a six‑layer architecture, vector‑based recall, various embedding and ranking models, multi‑task and multi‑scenario optimization techniques, and future directions for large‑model integration.

Deep LearningFAISSmulti-task learning
0 likes · 19 min read
Deep Learning Practices for Internet Real‑Estate Recommendation at 58.com
DataFunTalk
DataFunTalk
Mar 28, 2024 · Artificial Intelligence

Multi-Task and Multi-Scenario Algorithms for Recommendation Systems: Methods, Challenges, and Applications

This article presents a comprehensive overview of multi‑task and multi‑scenario recommendation algorithms, detailing background challenges, algorithm classifications such as TAML, CausalInt, and DFFM, their modular designs, experimental validations, and practical Q&A insights for large‑scale advertising systems.

Recommendation Systemsadvertising algorithmsmachine learning
0 likes · 19 min read
Multi-Task and Multi-Scenario Algorithms for Recommendation Systems: Methods, Challenges, and Applications
NetEase Smart Enterprise Tech+
NetEase Smart Enterprise Tech+
Feb 28, 2024 · Artificial Intelligence

Mastering Multi-Task Learning: Network Designs & Loss Balancing

This article reviews the challenges of multi‑task learning, compares various network architectures such as hard‑parameter sharing, MMoE, CGC, and PLE, and examines loss‑balancing techniques like GradNorm, Dynamic Weight Average and task‑prioritization, offering insights on how to mitigate the “seesaw” effect and improve overall performance.

AI researchNeural Networksdynamic weighting
0 likes · 15 min read
Mastering Multi-Task Learning: Network Designs & Loss Balancing
DataFunSummit
DataFunSummit
Feb 9, 2024 · Artificial Intelligence

STAN: A User‑Lifecycle‑Based Multi‑Task Recommendation Model for Shopee

The article introduces STAN, a multi‑task recommendation framework that leverages user lifecycle segmentation to jointly optimize CTR, stay‑time, and CVR, detailing the business context, key challenges, solution architecture, offline and online evaluations, and future research directions.

CTRCVRDeep Learning
0 likes · 8 min read
STAN: A User‑Lifecycle‑Based Multi‑Task Recommendation Model for Shopee
DataFunSummit
DataFunSummit
Dec 27, 2023 · Artificial Intelligence

Two-Stage Constrained Actor-Critic for Short‑Video Recommendation and a Reinforcement‑Learning Multi‑Task Framework

This article presents a two‑stage constrained actor‑critic (TSCAC) algorithm that models short‑video recommendation as a constrained reinforcement‑learning problem, details its theoretical formulation and optimization loss, and validates its superiority through extensive offline and online experiments, followed by a multi‑task reinforcement‑learning framework (RMTL) that further improves multi‑objective recommendation performance.

Recommendation SystemsReinforcement Learningconstrained optimization
0 likes · 16 min read
Two-Stage Constrained Actor-Critic for Short‑Video Recommendation and a Reinforcement‑Learning Multi‑Task Framework
Architect
Architect
Dec 14, 2023 · Artificial Intelligence

How Multi‑Task Multi‑Scene Modeling Powers ZhiZhuan’s Search: Algorithms, Industry Practices, and Lessons

This article analyzes the challenges of multi‑task and multi‑scene recommendation for large‑scale C‑end services, reviews key academic and industry solutions such as Shared‑Bottom, MMoE, PLE, ESMM, LHUC, PEPNet, MTMS and HiNet, and details ZhiZhuan’s end‑to‑end architecture that achieved over 6% click‑through and 2% conversion improvements.

AI model architectureRecommendation SystemsZhiZhuan
0 likes · 15 min read
How Multi‑Task Multi‑Scene Modeling Powers ZhiZhuan’s Search: Algorithms, Industry Practices, and Lessons
DataFunTalk
DataFunTalk
Nov 27, 2023 · Artificial Intelligence

STAN: A User‑Lifecycle‑Aware Multi‑Task Recommendation Model for Shopee

This article introduces STAN, a user‑lifecycle‑aware multi‑task recommendation model proposed by Shopee that refines CTR, CVR, and stay‑time predictions by identifying and tracking user states, demonstrates offline gains on Shopee and public datasets, and reports online improvements in click‑through, dwell‑time, and order metrics.

CTRCVRRecommendation Systems
0 likes · 8 min read
STAN: A User‑Lifecycle‑Aware Multi‑Task Recommendation Model for Shopee
Sohu Tech Products
Sohu Tech Products
Nov 8, 2023 · Artificial Intelligence

Two‑Stage Constrained Actor‑Critic for Short‑Video Recommendation and a Reinforcement‑Learning Multi‑Task Recommendation Framework

The presentation introduces a two‑stage constrained actor‑critic algorithm that learns auxiliary policies for interaction signals before optimizing watch‑time under KL constraints, and a reinforcement‑learning multi‑task learning framework that models session‑level dynamics with adaptive multi‑critic weighting, both achieving significant offline and online gains in short‑video recommendation.

Recommendation SystemsReinforcement Learningactor-critic
0 likes · 16 min read
Two‑Stage Constrained Actor‑Critic for Short‑Video Recommendation and a Reinforcement‑Learning Multi‑Task Recommendation Framework
Baobao Algorithm Notes
Baobao Algorithm Notes
Oct 25, 2023 · Artificial Intelligence

How Mixed Data Shapes LLaMA SFT: Scaling Trends, Conflict Zones, and the DMT Remedy

This article investigates how mixing data from mathematical reasoning, code generation, and general instruction-following tasks influences supervised fine‑tuning of LLaMA models, revealing distinct scaling curves, resource‑dependent performance conflicts, and a two‑stage DMT strategy that mitigates catastrophic forgetting while boosting overall capability.

DMT StrategyLLaMAModel Fine‑tuning
0 likes · 14 min read
How Mixed Data Shapes LLaMA SFT: Scaling Trends, Conflict Zones, and the DMT Remedy
DataFunSummit
DataFunSummit
Oct 9, 2023 · Artificial Intelligence

Multi-Task and Multi-Scenario Algorithms for Recommendation Systems: Methods, Challenges, and Applications

This article presents a comprehensive overview of multi‑task and multi‑scenario algorithms applied to recommendation systems, covering background challenges, algorithm taxonomy, recent research, detailed model architectures such as TAML, CausalInt and DFFM, experimental results on public and private datasets, and a Q&A discussion.

AdvertisingRecommendation Systemsmachine learning
0 likes · 20 min read
Multi-Task and Multi-Scenario Algorithms for Recommendation Systems: Methods, Challenges, and Applications
Network Intelligence Research Center (NIRC)
Network Intelligence Research Center (NIRC)
Sep 21, 2023 · Artificial Intelligence

RVM: Real-Time High-Resolution Video Matting

The article reviews the paper "Robust High-Resolution Video Matting with Temporal Guidance", detailing a GRU‑based multi‑task network that achieves real‑time performance on 4K (76 FPS) and 1080p (104 FPS) video by leveraging temporal information and semantic segmentation.

GRUMobileNetV3Real-Time
0 likes · 5 min read
RVM: Real-Time High-Resolution Video Matting
Kuaishou Tech
Kuaishou Tech
Aug 11, 2023 · Artificial Intelligence

PEPNet: Parameter and Embedding Personalized Network for Multi‑Task Multi‑Domain Recommendation

The paper introduces PEPNet, a plug‑and‑play network that tackles the domain‑seesaw and task‑seesaw problems in multi‑scenario recommendation by using a gated personalization module (GateNU) together with embedding‑level (EPNet) and parameter‑level (PPNet) personalization, and demonstrates its superiority through extensive offline and online experiments on Kuaishou data.

Deep LearningEmbeddinggate network
0 likes · 11 min read
PEPNet: Parameter and Embedding Personalized Network for Multi‑Task Multi‑Domain Recommendation
Alimama Tech
Alimama Tech
Aug 9, 2023 · Artificial Intelligence

Highlights of Alibaba Mama Team Papers Accepted at CIKM 2023

Eight Alibaba Mama team papers accepted at CIKM 2023 present advances such as task‑specific bottom‑representation networks for recommendation, a unified GNN for multi‑scenario e‑commerce search, multi‑slot bid shading, consistency‑oriented pre‑ranking, bias‑mitigating CTR prediction, efficient progressive‑sampling self‑attention, delayed‑feedback conversion modeling, and hybrid contrastive multi‑scenario ad ranking.

AICTR predictionGraph Neural Network
0 likes · 13 min read
Highlights of Alibaba Mama Team Papers Accepted at CIKM 2023
DataFunTalk
DataFunTalk
Jul 18, 2023 · Artificial Intelligence

Travel Demand Prediction and Recommendation Optimization at Fliggy: Challenges, Algorithm Evolution, and Future Directions

This article presents Fliggy's work on user travel demand prediction, outlining the unique challenges of travel scenarios, the evolution of recall and ranking algorithms—including multi‑task learning, graph‑based models, and intention‑capture mechanisms—and discusses future research directions such as long‑sequence modeling and cross‑domain learning.

Recommendation Systemsgraph neural networksmachine learning
0 likes · 19 min read
Travel Demand Prediction and Recommendation Optimization at Fliggy: Challenges, Algorithm Evolution, and Future Directions
DataFunSummit
DataFunSummit
May 29, 2023 · Artificial Intelligence

Neuron‑level Shared Multi‑task Learning for Joint CTR and CVR Prediction

This article introduces a neuron‑level shared multi‑task learning framework that jointly estimates click‑through rate (CTR) and conversion rate (CVR), discusses the background and advantages of multi‑task learning, reviews classic shared‑bottom models, describes the proposed pruning‑based architecture, and presents experimental results demonstrating its effectiveness in large‑scale recommendation systems.

CTRCVRModel Pruning
0 likes · 11 min read
Neuron‑level Shared Multi‑task Learning for Joint CTR and CVR Prediction
Kuaishou Tech
Kuaishou Tech
Apr 29, 2023 · Artificial Intelligence

RMTL: A Reinforcement Learning Based Multi‑Task Learning Framework for Session‑Level Recommendation

The paper proposes RMTL, a reinforcement‑learning driven multi‑task learning framework that builds session‑level MDPs, trains a multi‑task actor‑critic network with dynamic loss weighting, and demonstrates significant AUC improvements over state‑of‑the‑art MTL recommendation models on public datasets.

Reinforcement Learningactor‑criticadaptive loss weighting
0 likes · 8 min read
RMTL: A Reinforcement Learning Based Multi‑Task Learning Framework for Session‑Level Recommendation
DaTaobao Tech
DaTaobao Tech
Apr 24, 2023 · Artificial Intelligence

Daily Good Shop: Two‑Stage Card Ranking and Multi‑Task Modeling for E‑commerce Recommendations

Daily Good Shop improves e‑commerce recommendations by first ranking products with long‑term user behavior models, assembling top items into cards, then ranking those cards using a shared‑bottom multi‑task network that jointly predicts click, subscription and lead‑IPV, and finally re‑ranking card sequences via beam‑search, yielding over 2 % more clicks, 34 % more subscriptions, 33 % more lead‑IPV and 22 % longer dwell time.

multi-task learningrankingrecommendation
0 likes · 11 min read
Daily Good Shop: Two‑Stage Card Ranking and Multi‑Task Modeling for E‑commerce Recommendations
Kuaishou Tech
Kuaishou Tech
Apr 24, 2023 · Artificial Intelligence

Divide‑and‑Conquer Embedding‑Based Retrieval with Prompt‑Based Multi‑Task Learning for Large‑Scale Recommendation

This paper identifies the trade‑off between simple and hard negatives in embedding‑based retrieval for recommendation, proposes a clustering‑based divide‑and‑conquer framework combined with prompt‑driven multi‑task learning to improve relevance, diversity, and fairness, and validates the approach through offline metrics, online A/B tests, and comparative experiments.

Embedding RetrievalPrompt Tuningapproximate nearest neighbor
0 likes · 9 min read
Divide‑and‑Conquer Embedding‑Based Retrieval with Prompt‑Based Multi‑Task Learning for Large‑Scale Recommendation
DataFunTalk
DataFunTalk
Apr 24, 2023 · Artificial Intelligence

Evolution of Large‑Scale Recommendation Models at Weibo: Technical Roadmap and Recent Advances

This article reviews the evolution of Weibo's large‑scale recommendation technology, covering the system's business scenarios, technical roadmap, recent large model iterations, multi‑task and multi‑scenario modeling, feature engineering, consistency between recall and ranking, and emerging techniques such as causal inference and graph methods.

Recommendation Systemscausal inferencegraph embeddings
0 likes · 18 min read
Evolution of Large‑Scale Recommendation Models at Weibo: Technical Roadmap and Recent Advances
Tongcheng Travel Technology Center
Tongcheng Travel Technology Center
Apr 14, 2023 · Artificial Intelligence

Multi‑Business Recommendation System for the Tongcheng App Home Page Waterfall Flow

This article describes the architecture, data processing, city‑intent modeling, resource recall strategies, and multi‑task ranking models—including PLE‑CGC and ESMM—used to improve click‑through and conversion rates of the Tongcheng travel app's homepage waterfall‑flow recommendation, and outlines experimental results and future optimization directions.

CTRCVRESMM
0 likes · 10 min read
Multi‑Business Recommendation System for the Tongcheng App Home Page Waterfall Flow
Meituan Technology Team
Meituan Technology Team
Mar 23, 2023 · Artificial Intelligence

HiNet: A Hierarchical Information Extraction Network for Multi-Scenario Multi-Task Learning in Recommendation Systems

HiNet, a hierarchical information extraction network combining a scenario‑aware attentive layer and a customized gate‑control layer, jointly learns shared and scenario‑specific representations for multiple recommendation tasks, delivering consistently higher CTR and CTCVR performance across six Meituan restaurant scenarios than strong baselines in both offline and online evaluations.

hierarchical networkmulti-task learningonline A/B testing
0 likes · 19 min read
HiNet: A Hierarchical Information Extraction Network for Multi-Scenario Multi-Task Learning in Recommendation Systems
DataFunTalk
DataFunTalk
Feb 1, 2023 · Artificial Intelligence

Kuaishou Recommendation System: Architecture, CTR Modeling, Multi‑Domain Multi‑Task Learning, and Long‑Short Term Behavior Modeling

This article presents a comprehensive overview of Kuaishou's large‑scale recommendation system, detailing its pipeline, unique characteristics, CTR model improvements, the PPNet personalization network, multi‑domain multi‑task framework, short‑ and long‑term behavior sequence modeling, and the challenges of handling billions of features and trillions of parameters.

AICTR modelbehavior sequence modeling
0 likes · 12 min read
Kuaishou Recommendation System: Architecture, CTR Modeling, Multi‑Domain Multi‑Task Learning, and Long‑Short Term Behavior Modeling
NetEase Cloud Music Tech Team
NetEase Cloud Music Tech Team
Jan 10, 2023 · Artificial Intelligence

Sentiment Classification and Topic Clustering for NetEase Cloud Music Comments

To boost NetEase Cloud Music’s comment handling, the authors combine active‑learning‑driven relabeling, domain‑specific MLM pretraining, contrastive‑learning‑based sample expansion, and multi‑task BERT sharing to raise sentiment‑classification precision and recall above 90 % and double moderation clean‑rate, while employing prompt‑generated story themes, IP‑focused classifiers, and hot‑word aggregation for effective short‑text topic clustering and scalable, theme‑aware distribution.

NLPSentiment Analysisactive learning
0 likes · 10 min read
Sentiment Classification and Topic Clustering for NetEase Cloud Music Comments
DataFunTalk
DataFunTalk
Dec 7, 2022 · Artificial Intelligence

Entire Space Delayed Feedback with Cross‑Task Knowledge Distillation (ESDC) for Multi‑Task E‑commerce Recommendation

This article presents Xiaomi’s e‑commerce recommendation research, addressing four key challenges—sample selection bias, data sparsity, delayed feedback, and knowledge inconsistency—by introducing the Entire Space Delayed Feedback with Cross‑Task Knowledge Distillation (ESDC) model, which combines causal inference, cross‑task distillation, twin networks, and uncertainty weighting to improve CVR prediction and achieve a 15% GMV lift over the baseline.

AICVRDelayed Feedback
0 likes · 11 min read
Entire Space Delayed Feedback with Cross‑Task Knowledge Distillation (ESDC) for Multi‑Task E‑commerce Recommendation
Hulu Beijing
Hulu Beijing
Dec 2, 2022 · Artificial Intelligence

How Disney+ Designs a Multi‑Task Video Search Ranking Model

This article explains the architecture of a video search ranking system that combines a deep encoding network, multi‑task expert networks, and a bias‑correction module to jointly optimize relevance, click‑through rate, and watch time for streaming platforms.

Bias CorrectionDeep Learningfeature engineering
0 likes · 15 min read
How Disney+ Designs a Multi‑Task Video Search Ranking Model
Meituan Technology Team
Meituan Technology Team
Nov 24, 2022 · Artificial Intelligence

Cross‑Lingual Structured Sentiment Analysis with Data Augmentation and Auxiliary Tasks

Meituan's Voice Interaction team tackled the lack of low‑resource language annotations and high optimization costs in SemEval‑2022 Task 10 by leveraging the cross‑lingual XLM‑RoBERTa model together with multi‑task learning and two data‑augmentation strategies, achieving first place in the zero‑shot subtask and second place in the monolingual subtask.

Cross-Lingual TransferStructured Sentiment AnalysisXLM-RoBERTa
0 likes · 25 min read
Cross‑Lingual Structured Sentiment Analysis with Data Augmentation and Auxiliary Tasks
Inke Technology
Inke Technology
Oct 27, 2022 · Artificial Intelligence

Scaling Card‑Based Social Matching with Multi‑Task AI Models and Efficient Backend

This article details the design and optimization of Jimu’s card‑based stranger‑social recommendation system, covering product background, gameplay flow, technical challenges in strategy and engineering, a multi‑task AI ranking model, vector recall improvements, and the resulting performance gains.

Vector Retrievalbackend optimizationmulti-task learning
0 likes · 20 min read
Scaling Card‑Based Social Matching with Multi‑Task AI Models and Efficient Backend
Tencent Advertising Technology
Tencent Advertising Technology
Aug 15, 2022 · Artificial Intelligence

Mixture of Virtual‑Kernel Experts (MVKE) for Multi‑Objective User Profile Modeling

At KDD 2022, Tencent Advertising introduced the Multi‑Virtual‑Kernel Experts (MVKE) model, a novel multi‑task architecture that replaces traditional dual‑tower structures with virtual‑kernel experts and a gating mechanism to jointly learn user preferences across multiple actions and topics, achieving superior offline and online performance.

AdvertisingMVKEmulti-task learning
0 likes · 13 min read
Mixture of Virtual‑Kernel Experts (MVKE) for Multi‑Objective User Profile Modeling
NetEase Yanxuan Technology Product Team
NetEase Yanxuan Technology Product Team
Aug 15, 2022 · Artificial Intelligence

Evolution of the First-Focus Personalized Recommendation Model in E-commerce

The article details a step‑by‑step evolution of an e‑commerce platform’s top‑slot recommendation system, moving from a DCN‑mix single‑objective model through BST‑based dynamic features, position‑bias debiasing, multi‑task MMoE learning, and finally BST with target‑attention, each yielding measurable CTR, conversion, and user‑value gains.

CTR predictionmulti-task learningposition bias
0 likes · 22 min read
Evolution of the First-Focus Personalized Recommendation Model in E-commerce
JD Tech
JD Tech
Jul 21, 2022 · Artificial Intelligence

Improving JD Retail Recommendation Advertising Ranking with Variational Feature Learning, User Interest Network Optimization, and Global Collaborative Modeling

This article presents JD's comprehensive technical solution for boosting recommendation ad ranking by addressing cold‑start, shallow user interest extraction, and insufficient global data through a variational feature learning framework, enhanced user‑interest networks, and full‑domain collaborative modeling, achieving over 1% AUC gain and notable revenue growth.

CTR predictionDeep Learninge‑commerce
0 likes · 21 min read
Improving JD Retail Recommendation Advertising Ranking with Variational Feature Learning, User Interest Network Optimization, and Global Collaborative Modeling
JD Retail Technology
JD Retail Technology
Jun 27, 2022 · Artificial Intelligence

Advances in JD E‑commerce Advertising CTR Prediction: Variational Feature Learning, User Interest Network Optimization, and Global User Collaborative Modeling

This article presents JD's end‑to‑end improvements for advertising click‑through‑rate prediction, addressing cold‑start, deep user‑interest mining, and full‑domain collaborative information through a variational feature learning framework, enhanced interest networks (PPNet+, NeNet, Weighted‑MMoE) and exposure‑sequence modeling, achieving over 1% cumulative AUC gain and publication in top conferences.

CTR predictione-commerce recommendationmulti-task learning
0 likes · 21 min read
Advances in JD E‑commerce Advertising CTR Prediction: Variational Feature Learning, User Interest Network Optimization, and Global User Collaborative Modeling
DataFunSummit
DataFunSummit
Jun 8, 2022 · Artificial Intelligence

Search Term Recommendation: Scenarios, Algorithm Design, and Future Directions

This article presents a comprehensive overview of search term recommendation in QQ Browser, covering various recommendation scenarios, challenges, query library architecture, multi‑task ranking models, coarse‑to‑fine ranking pipelines, auto‑completion strategies, and future research directions.

AIRecommendation Systemsmachine learning
0 likes · 14 min read
Search Term Recommendation: Scenarios, Algorithm Design, and Future Directions
DataFunTalk
DataFunTalk
May 15, 2022 · Artificial Intelligence

Search Term Recommendation: Scenarios, Algorithm Design, Challenges and Future Directions

This article presents an in‑depth overview of search term recommendation in QQ Browser, covering the various recommendation scenarios, the composition of recommendation items, the multi‑stage algorithm architecture, key technical challenges, evaluation metrics, and future research directions such as multi‑task and session‑aware modeling.

future researchmachine learningmulti-task learning
0 likes · 15 min read
Search Term Recommendation: Scenarios, Algorithm Design, Challenges and Future Directions
DaTaobao Tech
DaTaobao Tech
Apr 26, 2022 · Artificial Intelligence

Optimization of Recall, Ranking, and Downward Modeling for the "Every Square Every House" Infinite-Scroll Light App

This article details a year‑long series of experiments on the Taobao “Every Square Every House” infinite‑scroll light app, describing how added recall paths, a coarse‑ranking filter, multi‑task MMOE sorting, a lightweight down‑scroll predictor, and relevance‑enhanced features together boosted click‑through, scroll depth and per‑user engagement by double‑digit percentages.

A/B testingModel Optimizationinfinite scroll
0 likes · 14 min read
Optimization of Recall, Ranking, and Downward Modeling for the "Every Square Every House" Infinite-Scroll Light App
Code DAO
Code DAO
Apr 24, 2022 · Artificial Intelligence

How Transfer Learning Accelerates Deep Learning Across Vision, NLP, and Reinforcement Learning

The article explains how transfer learning reduces data and time requirements in deep learning by reusing pretrained models for vision, natural language processing, and reinforcement learning, while discussing challenges such as overfitting, the need for progressive networks, entropy regularization, domain adaptation, multi‑task learning, and model distillation.

Deep LearningReinforcement Learningdomain adaptation
0 likes · 10 min read
How Transfer Learning Accelerates Deep Learning Across Vision, NLP, and Reinforcement Learning
Alimama Tech
Alimama Tech
Feb 23, 2022 · Artificial Intelligence

Meta‑Network Based Multi‑Scenario Multi‑Task Model (M2M) for Alibaba Advertising Merchants

The paper introduces a Meta‑Network based Multi‑Scenario Multi‑Task (M2M) model for Alibaba’s advertising merchants, combining a transformer‑driven backbone with scene‑aware meta‑learning modules to jointly predict spend, clicks and activity across diverse ad scenarios, achieving up to 27 % error reduction offline and over 2 % lifts in merchant activity and ARPU online.

AlibabaMeta LearningPrediction
0 likes · 14 min read
Meta‑Network Based Multi‑Scenario Multi‑Task Model (M2M) for Alibaba Advertising Merchants
DeWu Technology
DeWu Technology
Dec 27, 2021 · Artificial Intelligence

Multi-Objective Modeling and Practice in DeWu Community Recommendation System

DeWu Community’s recommendation system progressed from single‑objective CTR modeling to a multi‑objective framework that combines independent models for dwell time, video completion and user interactions via score‑fusion, ranking‑learning and multi‑task architectures with shared parameters and gradient‑blocking, delivering higher engagement and retention.

CTRModel Fusionmulti-task learning
0 likes · 15 min read
Multi-Objective Modeling and Practice in DeWu Community Recommendation System
DataFunTalk
DataFunTalk
Dec 18, 2021 · Artificial Intelligence

Adaptive Mutual Supervision Multi‑Task Graph Neural Network for Fine‑Grained Urban Traffic Demand Prediction

This work proposes an adaptive mutual‑supervision multi‑task graph neural network that captures spatio‑temporal dynamics and heterogeneous group behaviors to predict fine‑grained urban travel demand, demonstrating over 10% performance gains on real‑world Beijing and Shanghai datasets compared with classic baselines.

Deep LearningGraph Neural NetworkTraffic Prediction
0 likes · 24 min read
Adaptive Mutual Supervision Multi‑Task Graph Neural Network for Fine‑Grained Urban Traffic Demand Prediction
DataFunSummit
DataFunSummit
Dec 18, 2021 · Artificial Intelligence

Adaptive Mutual Supervision Multi‑Task Graph Neural Network for Fine‑Grained Urban Traffic Demand Prediction

This work introduces a novel adaptive mutual‑supervision multi‑task graph neural network that captures spatio‑temporal dynamics and group‑specific travel patterns, achieving over 10% improvement in short‑term traffic demand forecasts across heterogeneous urban populations.

Graph Neural Networkadaptive supervisionmulti-task learning
0 likes · 22 min read
Adaptive Mutual Supervision Multi‑Task Graph Neural Network for Fine‑Grained Urban Traffic Demand Prediction
58 Tech
58 Tech
Nov 25, 2021 · Artificial Intelligence

Technical Evolution of the “Guess You Want” Recommendation Module in 58 Local Services

This article describes the design, multi‑stage recall strategies, and successive ranking model upgrades—including BERT‑based intent prediction, vector‑based DSSM recall, tag expansion, and multi‑task DeepFM/MMoE/ESMM architectures—that together reduce no‑result rates and significantly improve user conversion for 58's local service platform.

BERTDSSMmulti-task learning
0 likes · 16 min read
Technical Evolution of the “Guess You Want” Recommendation Module in 58 Local Services
Meituan Technology Team
Meituan Technology Team
Nov 18, 2021 · Artificial Intelligence

Multi‑Business Product Ranking in Meituan Search: Challenges, Modeling Approaches, and Practical Results

Meituan Search tackles the difficulty of ranking items from diverse business lines by introducing a five‑tower mixed architecture, group‑lasso and feature‑gate selection, a probabilistic graph model, and a joint block‑order/size predictor, achieving notable offline NDCG gains and online CTR and purchase‑rate improvements.

Deep Learninge‑commercefeature selection
0 likes · 19 min read
Multi‑Business Product Ranking in Meituan Search: Challenges, Modeling Approaches, and Practical Results
Alimama Tech
Alimama Tech
Nov 10, 2021 · Artificial Intelligence

AutoHERI: Hierarchical Representation Automatic Aggregation for CVR Estimation in Advertising

AutoHERI, a hierarchical representation automatic aggregation model discovered via one‑shot neural architecture search, jointly learns CTR and CVR (and other downstream tasks) to capture cascade relationships, achieving superior AUC and conversion‑rate lifts in large‑scale Alibaba advertising datasets and prompting full production deployment.

AutoMLCVR estimationNeural Architecture Search
0 likes · 15 min read
AutoHERI: Hierarchical Representation Automatic Aggregation for CVR Estimation in Advertising
DataFunTalk
DataFunTalk
Nov 10, 2021 · Artificial Intelligence

Learnable Index Structures for Large‑Scale Retrieval: Deep Retrieval Model and Training Methods

This article introduces ByteDance's Deep Retrieval (DR) framework, describing its learnable index structure that aligns embedding training with retrieval objectives, detailing the core model, structure‑loss training via EM and online EM algorithms, beam‑search serving, multi‑task learning, and practical insights from Q&A.

Beam SearchEM algorithmRecommendation Systems
0 likes · 11 min read
Learnable Index Structures for Large‑Scale Retrieval: Deep Retrieval Model and Training Methods
DataFunTalk
DataFunTalk
Nov 2, 2021 · Artificial Intelligence

Personalized Recommendation and Advertising Algorithms for E‑commerce: Business Overview, Recall and Ranking Optimization, Multi‑Task Modeling, and Future Directions

This article presents a comprehensive technical overview of JD.com’s e‑commerce recommendation and advertising systems, covering business scenarios, recall optimizations (profile and similarity‑based), multi‑task ranking improvements, sample weighting, multi‑model ensembles, PID‑based CPC control, conversion‑delay modeling, and the achieved performance gains and future research plans.

CTR optimizationRecommendation Systemse‑commerce
0 likes · 18 min read
Personalized Recommendation and Advertising Algorithms for E‑commerce: Business Overview, Recall and Ranking Optimization, Multi‑Task Modeling, and Future Directions
DataFunSummit
DataFunSummit
Sep 30, 2021 · Artificial Intelligence

Transfer Learning for Financial Risk Control: Theory, Methods, and Empirical Results

This article introduces the fundamentals of transfer learning, formalizes its theoretical foundations, and demonstrates how multi‑task learning and domain adaptation techniques can be applied to financial risk control to overcome label scarcity, distribution shift, and improve model performance.

Artificial Intelligencedomain adaptationfinancial risk control
0 likes · 17 min read
Transfer Learning for Financial Risk Control: Theory, Methods, and Empirical Results
DataFunTalk
DataFunTalk
Sep 27, 2021 · Artificial Intelligence

Transfer Learning for Financial Risk Control: Theory, Methods, and Empirical Evaluation

This article introduces the fundamentals of transfer learning, explains its theoretical foundations and formulas, and demonstrates how multi‑task learning and domain‑adaptation techniques are applied to financial risk‑control scenarios to overcome label scarcity, distribution shift, and model complexity challenges, presenting detailed experimental results and analysis.

Deep LearningModel Evaluationdomain adaptation
0 likes · 17 min read
Transfer Learning for Financial Risk Control: Theory, Methods, and Empirical Evaluation
DataFunSummit
DataFunSummit
Sep 2, 2021 · Artificial Intelligence

Multi‑Task Learning Models for Recommendation Systems: An Industrial Survey

This article surveys recent industrial multi‑task learning approaches for recommendation, covering models such as Alibaba's ESMM and ESM2, DUPN, Meituan's deep ranking, Google’s MMoE, YouTube’s multi‑objective system, Zhihu’s ranking, and summarizing their architectures, loss functions, and practical gains.

CTRCVRMMoE
0 likes · 15 min read
Multi‑Task Learning Models for Recommendation Systems: An Industrial Survey
Meituan Technology Team
Meituan Technology Team
Aug 12, 2021 · Artificial Intelligence

Adaptive Information Transfer Multi-task (AITM) Framework for Sequential User Conversion Modeling in Targeted Display Advertising

The Adaptive Information Transfer Multi‑task (AITM) framework integrates multi‑task learning with an attention‑based information‑transfer module to jointly model the sequential conversion chain in targeted display ads, mitigating class imbalance and boosting end‑to‑end user acquisition rates, as demonstrated by offline and online experiments.

AITMSequential Modelingconversion rate
0 likes · 16 min read
Adaptive Information Transfer Multi-task (AITM) Framework for Sequential User Conversion Modeling in Targeted Display Advertising
Sohu Tech Products
Sohu Tech Products
Aug 4, 2021 · Artificial Intelligence

Technical Summary of the 2021 Sohu Campus Text Matching Algorithm Competition

This article presents a comprehensive technical summary of the 2021 Sohu Campus Text Matching Algorithm Competition, detailing data characteristics, preprocessing strategies, tokenization choices, positional encoding methods, model architectures using relative encodings such as WoBERT and RoFormer, experimental results, and reflections on future improvements.

Model DesignNLPcompetition
0 likes · 9 min read
Technical Summary of the 2021 Sohu Campus Text Matching Algorithm Competition
DataFunTalk
DataFunTalk
Jul 17, 2021 · Artificial Intelligence

Multi-Objective Modeling for CRM Opportunity Smart Allocation: Iterative Deep Learning Solutions

This article describes the evolution of a multi‑objective deep‑learning framework for automatically assigning CRM opportunities to salespeople, detailing five model versions—from an XGBoost baseline with sample weighting to advanced PLE‑based architectures—while reporting offline and online performance gains in both call‑out and connection‑out conversion rates.

A/B testingCRMDeep Learning
0 likes · 33 min read
Multi-Objective Modeling for CRM Opportunity Smart Allocation: Iterative Deep Learning Solutions
Meituan Technology Team
Meituan Technology Team
Jul 8, 2021 · Artificial Intelligence

Multi-Business Ranking Modeling in Meituan Search

Meituan Search tackles the multi‑business ranking challenge by introducing a quota‑allocation model (MQM) and a series of precise ranking models (MBN) that progressively incorporate sub‑networks, multi‑task learning and transformer‑based behavior sequences, delivering consistent CTR and purchase‑rate gains across food, hotel, travel and other services while outlining future work on feature utilization, sample‑imbalance mitigation and multi‑objective optimization.

MeituanRecommendation Systemsmachine learning
0 likes · 15 min read
Multi-Business Ranking Modeling in Meituan Search
58 Tech
58 Tech
Jul 7, 2021 · Artificial Intelligence

Multi‑Objective Modeling for CRM Opportunity Allocation: Iterative Deep Learning Approaches

This article details the development and iterative optimization of multi‑task deep learning models—including XGBoost‑based baselines, MMoE, ESMM‑enhanced MMoE, PLE, and bias‑aware ranking—to simultaneously improve call‑out and connect‑out rates in a CRM opportunity distribution system, presenting offline gains and online deployment results for each version.

CRMModel Optimizationmulti-task learning
0 likes · 33 min read
Multi‑Objective Modeling for CRM Opportunity Allocation: Iterative Deep Learning Approaches
iQIYI Technical Product Team
iQIYI Technical Product Team
Jun 25, 2021 · Artificial Intelligence

Multi-Objective Optimization in Short Video Recommendation at iQIYI

iQIYI improves short‑video recommendation by applying multi‑objective optimization—weighting clicks by watch duration, fusing separate click and watch‑time models, employing multi‑task learning with ESMM/MMOE and Pareto‑guided PSO hyper‑parameter search—delivering 7%+ watch‑time growth, 20%+ interaction gains, and 1.5‑3% CTR lifts while planning cross‑scene learning and further model refinements.

Model FusionParticle Swarm Optimizationmulti-task learning
0 likes · 14 min read
Multi-Objective Optimization in Short Video Recommendation at iQIYI
JD Tech
JD Tech
Jun 17, 2021 · Artificial Intelligence

MTrajRec: Map-Constrained Trajectory Recovery via Seq2Seq Multi‑Task Learning

The paper introduces MTrajRec, a Seq2Seq multi‑task learning framework that simultaneously restores low‑sampling‑rate GPS trajectories to high‑sampling‑rate and aligns them to the road network, achieving more accurate and efficient trajectory recovery for downstream applications such as navigation and travel‑time estimation.

Deep LearningKDD 2021Seq2Seq
0 likes · 9 min read
MTrajRec: Map-Constrained Trajectory Recovery via Seq2Seq Multi‑Task Learning
Didi Tech
Didi Tech
Jun 4, 2021 · Artificial Intelligence

Graph Convolutional Network for Shared Bike Demand Forecasting: Time Series Modeling and Multi‑Task Learning

The paper presents a graph convolutional network approach that leverages multi‑task learning and spectral graph convolutions to forecast shared‑bike inflow, outflow, and demand gaps across a city’s non‑Euclidean parking network, demonstrating improved accuracy over traditional time‑series baselines while noting scalability and directional graph limitations.

Demand ForecastingGCNGraph Neural Network
0 likes · 13 min read
Graph Convolutional Network for Shared Bike Demand Forecasting: Time Series Modeling and Multi‑Task Learning
DataFunTalk
DataFunTalk
May 9, 2021 · Artificial Intelligence

Few-Shot Learning, Data Augmentation, and Multi‑Task Learning for Safety Modeling in Ride‑Hailing Platforms

This article presents Didi's exploration of few‑shot learning, data‑augmentation, semi‑supervised self‑training and multi‑task learning techniques to address the scarcity of labeled samples in safety and governance scenarios, demonstrating practical solutions and performance gains across various risk‑detection tasks.

AIFew‑Shot LearningSemi-supervised Learning
0 likes · 15 min read
Few-Shot Learning, Data Augmentation, and Multi‑Task Learning for Safety Modeling in Ride‑Hailing Platforms
DataFunTalk
DataFunTalk
Apr 22, 2021 · Artificial Intelligence

Governance Algorithms for O2O Platforms: Challenges, Framework, and Model Exploration

This article presents Didi's comprehensive governance algorithm system for O2O platforms, detailing business background, technical challenges, a three‑stage algorithmic framework, model innovations such as small‑sample learning, multi‑task and transfer learning, and extensive feature engineering including multimodal and streaming features.

O2O platformsfeature engineeringgovernance algorithms
0 likes · 15 min read
Governance Algorithms for O2O Platforms: Challenges, Framework, and Model Exploration
Didi Tech
Didi Tech
Apr 20, 2021 · Artificial Intelligence

Few-Shot Learning, Data Augmentation, and Semi‑Supervised Methods for Improving Safety and Governance Models at Didi

To overcome scarce labeled data for safety and governance, Didi combines few‑shot learning with systematic data augmentation, self‑training semi‑supervised labeling, and multi‑task neural architectures, cutting labeling costs and reducing log‑loss by over 20% while boosting ROC‑AUC and PR‑AUC across harassment detection, expense‑complaint, and route‑intercept use cases.

AI SafetyDidiFew‑Shot Learning
0 likes · 15 min read
Few-Shot Learning, Data Augmentation, and Semi‑Supervised Methods for Improving Safety and Governance Models at Didi
Didi Tech
Didi Tech
Apr 16, 2021 · Artificial Intelligence

Governance Algorithms for O2O Ride-Hailing Platforms: Challenges, Framework, and Model Exploration

The paper presents Didi’s three‑layer governance‑algorithm framework for O2O ride‑hailing, addressing high business complexity, limited labeled data, interpretability, and multimodal features through small‑sample, transfer, and multi‑task learning, achieving notable gains in dispute resolution, NPS and CPO while highlighting remaining data and robustness challenges.

Ride Hailingfeature engineeringgovernance algorithms
0 likes · 15 min read
Governance Algorithms for O2O Ride-Hailing Platforms: Challenges, Framework, and Model Exploration
DataFunTalk
DataFunTalk
Mar 17, 2021 · Artificial Intelligence

Deep Ranking Model Evolution and Applications in Taobao Live: DBMTL, DMR, and RUI Ranking

This article presents a comprehensive overview of Taobao Live's deep ranking system evolution, detailing the DBMTL multi‑task learning framework, the two‑tower DMR matching‑ranking architecture, and the RUI Ranking refer‑item model, together with their offline formulas, online deployment scenarios, and measured performance gains across click‑through, watch‑time, and conversion metrics.

AIDeep LearningModel Optimization
0 likes · 27 min read
Deep Ranking Model Evolution and Applications in Taobao Live: DBMTL, DMR, and RUI Ranking
DataFunSummit
DataFunSummit
Mar 7, 2021 · Artificial Intelligence

A Comprehensive Overview of Multi‑Task Learning in AI: Concepts, Applications, and Practical Tips

This article provides an in‑depth introduction to multi‑task learning (MTL), explaining its core concepts, why it is widely used in recommendation systems, NLP, CV and reinforcement learning, and offering guidance on model architectures, loss design, auxiliary tasks, and practical deployment tips.

MTLNLPRecommendation Systems
0 likes · 19 min read
A Comprehensive Overview of Multi‑Task Learning in AI: Concepts, Applications, and Practical Tips
DataFunTalk
DataFunTalk
Feb 10, 2021 · Artificial Intelligence

Deep Learning Based Search Ranking Optimization for 58.com Rental Services

This article describes how 58.com’s rental platform leverages deep learning models such as Wide&Deep, DeepFM, DCN, DIN, and DIEN to improve search ranking, detailing data pipelines, feature engineering, model iteration, multi‑task training, prediction optimizations, and resulting online performance gains.

Deep LearningModel OptimizationRecommendation Systems
0 likes · 27 min read
Deep Learning Based Search Ranking Optimization for 58.com Rental Services
58 Tech
58 Tech
Jan 25, 2021 · Artificial Intelligence

Deep Learning Ranking Models for 58.com Rental Search: Architecture, Model Iterations, and Optimization

This article presents the end‑to‑end design, feature engineering, model evolution (Wide&Deep, DeepFM, DCN, DIN, DIEN), multi‑task training, and deployment optimizations that 58.com applied to improve search ranking for its rental business, demonstrating significant gains in click‑through and conversion rates.

Model Optimizationfeature engineeringmulti-task learning
0 likes · 28 min read
Deep Learning Ranking Models for 58.com Rental Search: Architecture, Model Iterations, and Optimization
DataFunTalk
DataFunTalk
Oct 22, 2020 · Artificial Intelligence

Analyzing Video Excitement: Methods, Frameworks, and Applications

This article presents a comprehensive overview of video excitement analysis, covering quality, aesthetics, and narrative factors, describing a multimodal framework with supervised, weakly supervised, and multi‑task models, and illustrating practical applications such as preview generation, clipping, and automatic cover creation.

Multimodal AIWeak Supervisioncontent recommendation
0 likes · 14 min read
Analyzing Video Excitement: Methods, Frameworks, and Applications
iQIYI Technical Product Team
iQIYI Technical Product Team
Sep 18, 2020 · Artificial Intelligence

iCartoonFace: A Large-Scale Cartoon Face Recognition Dataset and Multi‑Task Learning Framework

The paper presents iCartoonFace, the largest manually annotated cartoon‑face dataset with 5,013 identities and 389,678 images, and a multi‑task learning framework that jointly trains on cartoon and real faces using classification, unknown‑identity rejection, and domain‑adaptation losses, achieving state‑of‑the‑art recognition despite pose, occlusion, and illumination challenges.

domain adaptationface recognitionmulti-task learning
0 likes · 10 min read
iCartoonFace: A Large-Scale Cartoon Face Recognition Dataset and Multi‑Task Learning Framework
DataFunTalk
DataFunTalk
Aug 28, 2020 · Artificial Intelligence

Intelligent Traffic Distribution in 58 Local Services: Algorithmic Practices and System Optimization

This article presents a comprehensive overview of 58 Local Services' traffic distribution system, detailing the ecosystem, user interaction flow, challenges such as information homogeneity and complex user structures, and the algorithmic solutions—including information and knowledge structuring, multi‑task user intent modeling, layered optimization, and system integration—used to improve recall, ranking, and real‑time personalization.

AISearchinformation structuring
0 likes · 21 min read
Intelligent Traffic Distribution in 58 Local Services: Algorithmic Practices and System Optimization
Youku Technology
Youku Technology
Jul 30, 2020 · Artificial Intelligence

Key Technologies for Entertainment Content Flow Management: Multi-Task Guarantee Optimization Algorithm Practice

The presentation explains how a multi‑task guarantee optimization algorithm—illustrated with Youku’s new hot series—builds an exposure‑sensitivity model to allocate limited video‑placement resources across homepage and channel slots, overcoming manual rule limitations and simultaneously maximizing play counts while satisfying diverse scenario and content objectives.

Content OptimizationKDD2020algorithm
0 likes · 2 min read
Key Technologies for Entertainment Content Flow Management: Multi-Task Guarantee Optimization Algorithm Practice
iQIYI Technical Product Team
iQIYI Technical Product Team
Jul 10, 2020 · Artificial Intelligence

Video Highlight Analysis Technology Framework

iQIYI’s video highlight analysis framework combines a large supervised dataset, deep label distribution learning, multi‑task training with a canonical‑correlated autoencoder, and a weakly supervised ranking model enhanced by confidence weighting and graph convolution, then fuses these signals to improve highlight detection accuracy.

Weak Supervisiongraph convolutional networksmulti-task learning
0 likes · 17 min read
Video Highlight Analysis Technology Framework
DataFunTalk
DataFunTalk
May 29, 2020 · Artificial Intelligence

Model‑Independent Learning: Multi‑Task Learning and Transfer Learning

This article explains two model‑independent learning paradigms—multi‑task learning and transfer learning—detailing their motivations, sharing mechanisms, training procedures, theoretical formulations, and practical benefits such as improved generalization, data efficiency, and domain‑invariant representations.

Deep Learningdomain adaptationmachine learning
0 likes · 21 min read
Model‑Independent Learning: Multi‑Task Learning and Transfer Learning
Huajiao Technology
Huajiao Technology
Apr 7, 2020 · Artificial Intelligence

How Huajiao Live Built a From‑Scratch Personalized Recommendation System

This article analyzes Huajiao Live's end‑to‑end recommendation pipeline, covering basic concepts, recall and ranking algorithms—including collaborative filtering, matrix factorization, deep learning models—and multi‑objective optimization, while detailing the engineering workflow for training, deployment, and real‑time serving in a live‑streaming environment.

AIDeep Learningcollaborative filtering
0 likes · 17 min read
How Huajiao Live Built a From‑Scratch Personalized Recommendation System
Huajiao Technology
Huajiao Technology
Jan 21, 2020 · Artificial Intelligence

Overview of Ranking Algorithms in Recommendation Systems

This article reviews the evolution of ranking models in modern recommendation systems, covering traditional linear models, factorization machines, tree‑based GBDT+LR, and a range of deep learning architectures such as Wide&Deep, DeepFM, DCN, xDeepFM, DIN, as well as multi‑task frameworks like ESMM and MMOE, and finally illustrates their practical deployment in a live streaming platform.

Deep LearningRecommendation Systemsfeature engineering
0 likes · 20 min read
Overview of Ranking Algorithms in Recommendation Systems
DataFunTalk
DataFunTalk
Dec 30, 2019 · Artificial Intelligence

Technical Trends in Recommendation Systems: From Retrieval to Re‑ranking

This article surveys recent advances in recommendation system technology, covering the evolution from a two‑stage recall‑ranking pipeline to a four‑stage architecture, and detailing emerging trends in model‑based recall, user‑behavior sequence modeling, knowledge‑graph integration, graph neural networks, advanced ranking models, multi‑objective optimization, multimodal fusion, and listwise re‑ranking.

Recommendation Systemsgraph neural networksinformation retrieval
0 likes · 45 min read
Technical Trends in Recommendation Systems: From Retrieval to Re‑ranking
58 Tech
58 Tech
Nov 29, 2019 · Artificial Intelligence

Ranking Strategy Optimization Practices for Commercial Traffic at 58.com

This article details the end‑to‑end optimization of 58.com’s commercial traffic ranking system, covering data‑flow upgrades, advanced feature engineering, real‑time and multi‑task model improvements, and a multi‑factor ranking mechanism, while sharing practical results and future directions.

Real-time Data Pipelinefeature engineeringmachine learning
0 likes · 17 min read
Ranking Strategy Optimization Practices for Commercial Traffic at 58.com
DataFunTalk
DataFunTalk
Oct 16, 2019 · Artificial Intelligence

Deep Learning Practices for Personalized Recommendation at Meitu: From Recall to Ranking

This article details Meitu's large‑scale personalized recommendation pipeline, describing the business scenario, challenges of massive data, latency and long‑tail distribution, and the application of deep learning techniques such as Item2vec, YouTubeNet, dual‑tower DNN, NFM, NFwFM and multi‑task learning to improve click‑through rate, conversion and user engagement.

Deep LearningRecommendation Systemslarge scale
0 likes · 20 min read
Deep Learning Practices for Personalized Recommendation at Meitu: From Recall to Ranking
DataFunTalk
DataFunTalk
Sep 27, 2019 · Artificial Intelligence

Applying Deep Learning to Meitu Community Recommendation: Embedding, Recall, and Ranking Models

The talk by Meitu senior algorithm expert Chen Wenqiang details how deep‑learning‑driven embedding, recall, and ranking techniques—including Item2vec, twin‑tower DNNs, and multi‑task NFwFM—are applied to improve click‑through rates, follow conversions, and user engagement in Meitu's content community.

AIDeep LearningRecommendation Systems
0 likes · 3 min read
Applying Deep Learning to Meitu Community Recommendation: Embedding, Recall, and Ranking Models
DataFunTalk
DataFunTalk
Jul 23, 2019 · Artificial Intelligence

Technical Exploration of Intelligent Dialogue Robots in Didi Ride-Hailing Scenarios

The talk presents Didi AI Labs' research on intelligent dialogue robots for ride‑hailing, covering single‑turn QA, multi‑turn conversation, multi‑task learning architectures, model experiments, active learning pipelines, and the overall system design that integrates intent detection, slot extraction, dialogue management, and response generation.

AIBERTDialogue Systems
0 likes · 10 min read
Technical Exploration of Intelligent Dialogue Robots in Didi Ride-Hailing Scenarios
DataFunTalk
DataFunTalk
Dec 28, 2018 · Artificial Intelligence

Zhihu Recommendation Page Ranking: Architecture, Feature Engineering, Model Evolution, and Future Directions

This article presents a comprehensive overview of Zhihu's recommendation page ranking system, covering its request flow, historical ranking evolution, feature design, deep learning models, multi‑task CTR optimization, practical engineering insights, current challenges, and future research directions such as reinforcement learning.

CTRmulti-task learningranking
0 likes · 15 min read
Zhihu Recommendation Page Ranking: Architecture, Feature Engineering, Model Evolution, and Future Directions
Alibaba Cloud Developer
Alibaba Cloud Developer
Aug 17, 2018 · Artificial Intelligence

Can Multi‑Task Learning Shorten E‑Commerce Titles Without Losing Sales?

This paper proposes a multi‑task learning approach that compresses overly long e‑commerce product titles into concise short titles using a Pointer Network, while simultaneously generating user search queries with an attention‑based encoder‑decoder, achieving higher readability, informativeness, and conversion rates than traditional methods.

Attention MechanismSequence-to-Sequencee-commerce SEO
0 likes · 11 min read
Can Multi‑Task Learning Shorten E‑Commerce Titles Without Losing Sales?