Kuaishou's Accepted Papers at ICLR 2025 and Their Summaries

The article highlights Kuashou's eleven high‑quality papers accepted at ICLR 2025, covering advances in streaming video understanding, 3D trajectory control, multimodal talking‑face animation, transformer indexing, efficient video generation, industrial recommendation datasets, token gradient conflict in MoE, stable segmentation, multi‑camera video synthesis, large‑scale multimodal instruction tuning, and hallucination detection in retrieval‑augmented generation.

AIResearchDeepLearningICLR2025

0 likes · 20 min read

Kuaishou's Accepted Papers at ICLR 2025 and Their Summaries

Meituan Technology Team

Nov 9, 2023 · Artificial Intelligence

Deep Contextual Interest Network (DCIN) for CTR Prediction

This article introduces the Deep Contextual Interest Network (DCIN), a novel CTR prediction model that jointly models clicked items, surrounding display context, and position bias through three modules—PCAM, FCFM, and IMM—showing significant offline AUC gains and a 1.5% online CTR improvement.

ABTestingCTRContextualModeling

0 likes · 22 min read

Deep Contextual Interest Network (DCIN) for CTR Prediction

DataFunSummit

Feb 11, 2023 · Artificial Intelligence

FiBiNET and FiBiNET++: Feature Importance and Bilinear Interaction for Click‑Through Rate Prediction

The article introduces FiBiNET, a CTR prediction model that incorporates a SENet module for dynamic feature‑importance learning and a bilinear‑interaction layer for enhanced second‑order feature interactions, then details its improved variant FiBiNET++ which reduces parameters with Bi‑Linear+ and an enhanced SENet+.

BilinearInteractionCTRDeepLearning

0 likes · 8 min read

FiBiNET and FiBiNET++: Feature Importance and Bilinear Interaction for Click‑Through Rate Prediction

Alibaba Cloud Developer

Sep 26, 2018 · Artificial Intelligence

How Alibaba’s DeepInsight Platform Transforms Deep Learning Model Debugging

DeepInsight, Alibaba’s deep‑learning quality platform, centralizes training data, visualizes intermediate metrics, and offers advanced debugging tools—including a revamped TensorBoard+, TensorViewer log management, enhanced evaluation metrics, and research‑driven insights—helping engineers pinpoint model issues and accelerate optimization in complex AI workflows.

AIPlatformDataVisualizationDeepLearning

0 likes · 9 min read

How Alibaba’s DeepInsight Platform Transforms Deep Learning Model Debugging