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.
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.
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.
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.
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.
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.
This article reviews a 2017 ACM MM paper that introduces a multi‑level similarity‑aware CNN (MSP‑CNN) for person re‑identification, detailing its siamese architecture, dual similarity constraints, multi‑task training, experimental results on CUHK03, Market‑1501 and CUHK01, and its advantages for large‑scale deployment.
Researchers propose a multi‑task learning approach that compresses overly long e‑commerce product titles into concise short titles by jointly training a Pointer Network for extraction and an encoder‑decoder for query generation, preserving key information and maintaining conversion rates, as validated by offline and online experiments.
This paper introduces a novel multi‑level similarity‑aware CNN (MSP‑CNN) for person re‑identification, applying distinct similarity constraints to low‑ and high‑level feature maps, integrating classification and similarity losses in a multitask framework, and demonstrating superior performance on CUHK03, CUHK01 and Market‑1501 benchmarks.
This paper proposes a novel attention‑based framework that groups and encodes heterogeneous user behavior sequences into separate semantic subspaces, applies self‑attention to capture inter‑behavior influences, and demonstrates faster training and comparable or improved recommendation performance across multiple tasks and datasets.
The paper describes how Meituan tackles information overload on its lifestyle platform by training multi‑task deep neural networks on billions of interaction logs using a distributed PS‑Lite framework, employing sophisticated feature engineering, missing‑value imputation, KL‑regularization and Neural Factorization Machines to boost offline AUC and online CTR in the “Guess You Like” recommendation feed, while introducing training‑time optimizations and outlining future multi‑task and contextual enhancements.
