Tagged articles
GAN
70 articles · Page 1 of 1
This survey reviews 122 LSTM innovations categorized into internal modifications, structural enhancements, mechanism fusions, and emerging architecture combinations, highlighting achievements like 20x molecular activity improvement in drug discovery and 40% inference speed gains, demonstrating LSTM's continued relevance due to simplicity, robustness, and low resource demands.
CLMGANKAN
0 likes · 4 min read

The article exposes how many low‑priced chargers falsely marketed as “120W” deliver only 18‑22W, lack proper safety features, and often use proprietary protocols that prevent fast charging, urging buyers to verify output specs, choose original or certified accessories, and prefer GaN technology for reliable performance.
GANconsumer safetyfast charging
0 likes · 7 min read


Qborfy AIApr 19, 2026 · Artificial Intelligence
The article details how a GAN‑inspired multi‑agent harness—combining a generator, an evaluator, and a planner—overcomes context‑window anxiety and self‑evaluation bias, enabling Claude to produce higher‑quality front‑end designs and full‑stack applications through iterative scoring, sprint contracts, and extensive cost‑benefit experiments.
AI EngineeringFull-Stack DevelopmentGAN
0 likes · 19 min read

A paper co‑authored by Xiaomi’s mobile RF team, Nengxun High‑Energy Semiconductor and HKUST was selected for IEDM 2025, reporting the first integration of a silicon‑based GaN low‑voltage power‑amplifier MMIC into a handset with over 50% efficiency, marking a breakthrough for 6G‑era RF front‑ends.
GANIEDM 2025Low-Voltage PA
0 likes · 8 min read

This article systematically explores the technical evolution, core principles, and emerging innovations of AI‑generated video, covering generation methods, GAN and diffusion models, transformer‑based DiT architectures, efficiency‑boosting NCR, audio‑visual V2A integration, and real‑world applications across media, education, and commerce.
AI video generationGANNCR
0 likes · 25 min read

This article compiles the ten core questions from a Tencent algorithm interview, covering overfitting, regularization, generalization error, dropout, residual connections, attention, embeddings, BART vs BERT, instruction‑tuning data, LLM hallucination, and why GANs collapse more than diffusion models, with concise explanations and interview‑ready tips.
GANLLMRegularization
0 likes · 22 min read

This article outlines the four dominant generative AI architectures—Transformers, diffusion models, GANs, and RNNs—explaining their core mechanisms, key capabilities, and typical application domains such as chatbots, image creation, deep‑fake media, and time‑series analysis, helping readers choose the right model for their needs.
AI applicationsGANRNN
0 likes · 3 min read


NewBeeNLPNov 11, 2024 · Artificial Intelligence
MIT’s 6.S978 Deep Generative Models seminar, taught by Associate Professor He Kaiming, offers graduate students a 15‑week deep dive into VAEs, autoregressive models, GANs, diffusion techniques, and cross‑disciplinary applications, with detailed weekly topics, required assignments, and publicly available lecture PDFs.
Deep Generative ModelsGANHe Kaiming
0 likes · 5 min read

This article reviews the definition, key metrics, and major model families of AI‑generated content, details XiaoBu’s recent breakthroughs in image generation, and discusses open research problems such as evaluation gaps, transformer limitations, and the need for richer multimodal intelligence representations.
AIGCGANVAE
0 likes · 14 min read

This article provides a friendly, step‑by‑step overview of five core deep‑learning architectures—CNN, RNN, GAN, Transformers, and encoder‑decoder—explaining their structures, key components, and typical use cases in image and natural‑language processing.
CNNDeep LearningEncoder-Decoder
0 likes · 12 min read

The survey traces the evolution of image‑quality evaluation for text‑to‑image generation—from early handcrafted edge and color cues, through GAN‑era similarity scores such as IS, FID and KID, to modern perceptual and CLIP‑based metrics like LPIPS, CLIPScore, TRIQ, IQT and human‑preference models—highlighting a shift toward semantic, aesthetic, and text‑image alignment measures and forecasting domain‑specific metrics for future diffusion models.
Evaluation MetricsGANgenerative models
0 likes · 18 min read

This comprehensive review examines the rapid advances in generative AI for multimodal image synthesis and editing, covering visual, textual, and audio guidance, model families such as GANs, diffusion, autoregressive, and NeRF, as well as datasets, challenges, and future research directions.
GANGenerative AINeRF
0 likes · 6 min read

The Kuaishou audio team secured first place in both the ICASSP 2024 Speech Signal Improvement and Audio Deep Packet Loss Concealment challenges by deploying a two‑stage GAN‑based speech enhancement system and a hybrid time‑frequency packet‑loss concealment model that dramatically improve real‑time communication quality.
GANICASSP 2024audio processing
0 likes · 8 min read

Google, in collaboration with UC Berkeley, has unveiled a novel generative AI approach called the Idempotent Generative Network (IGN) that can produce images from any input in a single step, offering an alternative to GANs, diffusion models, and consistency models.
GANGenerative AIIGN
0 likes · 3 min read

The article analyzes the limitations of traditional speech‑driven lip‑sync methods and explains how Wav2Lip introduces a pretrained multi‑frame expert sync discriminator, a two‑stage GAN training pipeline, and a specialized generator architecture to produce high‑quality, audio‑aligned facial videos.
Computer VisionDeep LearningGAN
0 likes · 7 min read

Generative AI, a branch of artificial intelligence that creates novel content such as text, images, and music, works by learning patterns from training data, with common models including GANs, VAEs, autoregressive and Transformer-based architectures, and its development involves task definition, data preparation, model design, training, evaluation, and ethical considerations.
GANGenerative AITransformer
0 likes · 8 min read

This article provides a comprehensive overview of deep neural network families, introducing twelve major architectures—including Feedforward, CNN, RNN, LSTM, DBN, GAN, Autoencoder, Residual, Capsule, Transformer, Attention, and Deep Reinforcement Learning—explaining their principles, structures, training methods, and offering Python/TensorFlow/PyTorch code examples.
CNNDeep LearningGAN
0 likes · 29 min read

This article presents NetShare, an end-to-end framework that uses time‑series GANs combined with domain‑specific encoding to synthesize privacy‑preserving IP header and flow traces, achieving up to 46% higher accuracy than prior generative baselines while improving the fidelity‑privacy trade‑off.
GANIP Header TracingNetShare
0 likes · 7 min read

This article reviews eight CVPR 2023 papers selected by Meituan’s technology team, covering self‑supervised learning, domain adaptation, federated learning, object detection, 3D reconstruction, GAN‑based pre‑training, RGB‑T tracking, vision‑language navigation, and visual‑textual layout generation, highlighting each work’s methodology, experiments, and reported performance gains.
3D Object DetectionCVPR 2023Computer Vision
0 likes · 15 min read

Alibaba Mama’s team introduces PDA‑GAN, an unsupervised domain‑adaptation framework employing a lightweight pixel‑level discriminator to align repaired and original image features, enabling image‑aware layout generation that outperforms prior methods on visual‑quality and layout metrics for advertising creatives.
GANlayout generationpixel-level discriminator
0 likes · 10 min read

This article presents NIO Power's intelligent equipment health management solution, detailing business background, operational challenges, PHM difficulties, and frontier AI technologies such as GAN‑based unsupervised anomaly detection and Conceptor‑based small‑sample fault diagnosis, illustrated with real‑world case studies and a comprehensive Q&A.
ConceptorGANNIO Power
0 likes · 28 min read

The article surveys AI‑generated art, explaining how GANs’ limitations gave way to diffusion models and the open‑source Stable Diffusion platform, which offers text‑to‑image, img2img, inpainting, DreamBooth fine‑tuning, and widespread commercial and DIY deployments via cloud or local WebUI setups.
AI artGANStable Diffusion
0 likes · 13 min read

This article introduces deep learning as a powerful AI technique, explains its core algorithms—including supervised, unsupervised, and semi‑supervised approaches—and provides concrete examples such as CNN, RNN, autoencoders, GAN, self‑supervised and transfer learning, illustrated with visual demos.
AIDeep LearningGAN
0 likes · 6 min read


AntTechDec 20, 2022 · Artificial Intelligence
Researchers from multiple institutions propose a GAN‑based video generation framework that explicitly models short‑, medium‑, and long‑range temporal relations, introduces B‑spline motion embeddings and temporal shift modules, and demonstrates substantial quality improvements across several video datasets.
B-splineGANStyleGAN-V
0 likes · 7 min read

This article explains the motivation behind Spectral Normalization for stabilizing GAN training, derives its mathematical foundation, shows how it enforces the Lipschitz constraint, and provides a complete PyTorch implementation with detailed code explanations.
Deep LearningGANLipschitz Continuity
0 likes · 11 min read


HelloTechOct 19, 2022 · Artificial Intelligence
The Intelligent Creative System defines advertising creatives across formats, evaluates image and text quality using reference‑based metrics and models like DeepBIQ, generates multimodal ads via GANs and Transformers, and selects optimal variants through bandit‑based CTR prediction and multimodal fusion, enabling scalable, data‑driven creative production.
AIBandit ModelGAN
0 likes · 10 min read

This article explains the GANomaly model for semi‑supervised anomaly detection, detailing its generator‑encoder‑discriminator architecture, loss functions, testing phase scoring, and provides annotated PyTorch source code to help readers implement and understand the approach.
Deep LearningEncoder-DecoderGAN
0 likes · 15 min read

This article explains why traditional GANs suffer from training instability, introduces the Wasserstein (Earth‑Mover) distance as a smoother alternative, derives the WGAN objective, discusses Lipschitz constraints, and provides practical PyTorch code modifications to convert a vanilla GAN into a stable WGAN.
Deep LearningGANPyTorch
0 likes · 21 min read

This article introduces the EGBAD model, an efficient GAN‑based anomaly detection method that replaces AnoGAN's costly latent variable search with an encoder, provides detailed PyTorch code for data loading, model construction, training, and inference, and compares its testing speed with AnoGAN.
DiscriminatorEGBADEncoder
0 likes · 18 min read

This talk explores the challenges and opportunities of leveraging sparsity in machine learning models, covering fine‑grained and coarse‑grained sparsity, NVIDIA Ampere’s 2:4 structured sparsity, knowledge‑distillation techniques for converting unstructured to structured sparsity, and model compression strategies for generative adversarial networks.
Deep LearningGANGPU acceleration
0 likes · 14 min read

This paper proposes a pix2pix‑based method to automatically generate green areas for large‑scale outdoor 3D scene modeling, detailing dataset creation via OpenCV segmentation, model training, region partitioning, and experimental results showing a 93.8% acceptance rate, significantly improving efficiency over manual drawing.
3D modelingComputer VisionGAN
0 likes · 14 min read

The paper introduces a perception‑driven automatic graphic layout system that aligns advertising creatives with product images via a domain‑alignment module and generates content‑aware layouts using a multi‑scale CNN‑Transformer generator, achieving superior aesthetic quality and lower element overlap compared with existing template‑based and deep‑learning methods.
GANImage Generationcontent-aware
0 likes · 10 min read


Code DAOJun 3, 2022 · Artificial Intelligence
This article walks through building and analyzing a minimal GAN—PointGAN—that learns to output the single value 1, covering the linear generator, a two‑layer discriminator, training loops, loss visualizations, instability diagnostics, and practical fixes such as loss easing, weighted examples, weight decay, and noisy generator parameters.
DiscriminatorGANNoise Injection
0 likes · 24 min read

This article presents a comprehensive overview of adversarial example research applied to captcha systems, covering the definition and history of adversarial attacks, geometric‑aware generation frameworks, FGSM‑based attack variants, experimental results, trade‑offs between image quality and attack strength, and future work such as AdvGAN integration.
AI safetyDeep LearningFGSM
0 likes · 14 min read

Alibaba Mama’s brand advertising value system evolves from instant CTR to deep CVR and causal uplift modeling, employing focal loss, multi‑task training, GAN‑based uplift, enriched user‑sequence and UID embeddings, which together improve conversion lift, QINI, and interaction metrics while mitigating exposure bias and delayed feedback.
CTRCVRGAN
0 likes · 16 min read

This article explains how Kuaishou Y‑Tech combines image inpainting, semantic editing, and advanced GAN techniques to accurately locate and remove facial wrinkles—especially neck wrinkles—while preserving realistic skin texture and achieving high‑quality results suitable for production deployment.
Computer VisionDeep LearningGAN
0 likes · 17 min read

The paper presents RISA, a reference‑guided image synthesis assessment model that learns to score the quality of a single generated image without human‑labeled data by leveraging GAN intermediate outputs, pixel‑wise interpolation, multiple binary classifiers, and contrastive learning, achieving results comparable to human perception and earning an AAAI 2022 oral presentation.
AIGANRISA
0 likes · 8 min read

This article introduces two AI-driven anime portrait tools—Magical Anime Portraits and the open‑source SofGAN project—explaining their workflow, underlying GAN technology, related research, and how they enable users to create unique, customizable anime avatars without manual drawing.
AIAnime AvatarGAN
0 likes · 6 min read


Code DAODec 11, 2021 · Artificial Intelligence
This article explains how to apply a Deep Convolutional GAN in PyTorch to create realistic synthetic images of marine plastic, addressing dataset scarcity, detailing the network architecture, training procedure, and showing loss curves and generated samples.
DCGANGANImage Generation
0 likes · 13 min read

iQIYI’s short‑video team solves the new‑video cold‑start problem by using a GAN that generates latent user features from video attributes and a discriminator to validate them, then matches these vectors to real users via cosine similarity, achieving double‑digit gains in exposure, CTR, and watch time.
Cold StartGANRecommendation System
0 likes · 13 min read

This article explains the GAN‑based “TuYa” sketch‑to‑landscape algorithm presented at the Yidian News Hackathon, detailing its semantic image synthesis approach, the encoder, generator with SPADE, and PatchGAN discriminator, and discusses potential applications for designers and architects.
AIComputer VisionGAN
0 likes · 9 min read

This article presents a concise collection of practical deep‑learning techniques—including cyclic learning‑rate, flooding, warmup, RAdam, adversarial training, focal loss, dropout, various normalization methods, ReLU, group normalization, label smoothing, Wasserstein GAN, skip connections, and weight initialization—along with code snippets and references for implementation.
Deep LearningGANRegularization
0 likes · 8 min read

The OSGAN method introduced by Alibaba’s Mama team and Prof. Song Ming‑Li merges generator and discriminator updates into a single stage, cutting GAN training time by roughly 1.5‑1.7× while maintaining performance, and is validated on symmetric and asymmetric DCGANs with open‑source code.
Computer VisionDeep LearningEfficiency
0 likes · 10 min read

This article reviews a GAN‑based deep learning method for removing motion blur from images, covering the problem definition, related work, the multi‑scale generator and discriminator architecture, loss functions, the GoPro dataset, and experimental results that demonstrate clear visual improvements.
Computer VisionDeep LearningGAN
0 likes · 11 min read

This article presents an overview of Baidu's video foundation technology architecture, covering the video R&D platform, core AI techniques for video understanding, editing, surveillance, and general vision, and detailing innovations such as Attention‑Cluster networks, cross‑modality attention with graph convolution, GANs, super‑resolution, and adaptive encoding.
Adaptive EncodingAttention MechanismGAN
0 likes · 14 min read

This article details Kuaishou Y‑Tech's practical deployment of GAN‑based high‑precision face attribute editing—covering gender, age, hair, and expression transformations—for short‑video effects, discussing background, business applications, technical challenges, and solutions across data preparation, model training, and mobile deployment.
Computer VisionGANKuaishou
0 likes · 15 min read

This tutorial shows how to use the open‑source PaddleGAN toolkit and its First‑Order Motion algorithm to generate custom "Ant yahei" face‑swap videos, covering installation, motion‑transfer commands, audio addition, and a glimpse of other powerful GAN features.
AIFirst Order MotionGAN
0 likes · 6 min read

This article provides a comprehensive overview of Generative Adversarial Networks, covering their original formulation, training dynamics, loss functions, major variants such as DCGAN and WGAN, and practical implementations for image‑to‑image translation, style transfer, and handwriting font synthesis at Laiye Technology.
Computer VisionDeep LearningGAN
0 likes · 28 min read

This article provides an in‑depth overview of Generative Adversarial Networks (GANs), covering their original formulation, major variants such as DCGAN and WGAN, challenges like mode collapse, image‑to‑image translation techniques (cGAN, pix2pix, CycleGAN), and practical handwriting style‑transfer implementations using BicycleGAN and Zi2Zi.
GANGenerative Adversarial Networksimage-to-image translation
0 likes · 27 min read

iQIYI’s AI‑driven coloring engine uses a GAN‑based model with attention and semantic reference modules to automate line‑art coloring, delivering consistent styles across frames, cutting coloring time to 0.7 seconds per image and reducing labor costs by about 30% in animation pipelines.
AIDeep LearningGAN
0 likes · 5 min read

At INTERSPEECH 2020 the authors introduced a phase‑aware music super‑resolution system that uses a frequency‑domain GAN combined with an enhanced Griffin‑Lim algorithm to reconstruct missing high‑frequency magnitude and phase, delivering brighter, louder, and more natural‑sounding recordings that surpass traditional interpolation and naive phase‑flipping methods.
GANINTERSPEECH 2020audio super-resolution
0 likes · 7 min read

Cartoonize is an open‑source web application that leverages a white‑box GAN model to convert images and short videos into high‑quality cartoon style, offering easy Docker or virtualenv installation, detailed usage instructions, and insights into the underlying research paper.
AIGANcartoonization
0 likes · 9 min read

The article examines how AI-driven mosaic removal tools, leveraging GAN technology, are reviving uncensored adult video content, sparking privacy, copyright, and ethical debates while detailing the technical challenges, industry reactions, and the rise of related software among Chinese users.
AIAdult VideoCopyright
0 likes · 8 min read

The article describes Ctrip's payment fraud risk characteristics, a comprehensive automated iterative anti‑fraud model framework—including variable system, GAN‑augmented sample generation, RNN behavior encoding, and tree‑based classifiers—and demonstrates how this approach restores recall performance compared with traditional static models.
Anti-FraudGANRNN
0 likes · 12 min read

This article introduces AnimeGAN, a lightweight GAN that converts real photos into Japanese anime‑style illustrations, explains its architecture, loss functions, model size advantages, and provides step‑by‑step instructions with code for setting up, training, and testing the TensorFlow implementation.
AnimeGANDeep LearningGAN
0 likes · 8 min read

This article provides a comprehensive overview of voice conversion technology, covering its definition, parallel and non‑parallel data approaches, classic and deep‑learning methods such as DTW, GMM, seq2seq, PPG, VAE, Flow, GAN, and practical applications and challenges in iQIYI’s products.
ASRDeep LearningGAN
0 likes · 8 min read

Voice conversion (VC) technology changes a speaker’s timbre and style while keeping the spoken text unchanged, supporting one‑to‑one, many‑to‑one, and many‑to‑many scenarios for medical assistance and entertainment, using parallel or non‑parallel data through methods such as DTW‑aligned frame mapping, attention‑based neural networks, PPG‑LSTM pipelines, VAEs, normalizing‑flow models, and GANs, with iQIYI focusing on non‑parallel data, prosody preservation, and noise‑robust augmentation.
Deep LearningGANVAE
0 likes · 12 min read

This article explains how Alibaba's Front‑End Intelligent project automatically converts UI design images into production‑ready code, covering layout analysis, background and foreground processing, a fusion of traditional image algorithms with deep‑learning detection, GAN‑based complex‑background extraction, experimental results and real‑world deployment.
GANautomationcode generation
0 likes · 21 min read

The proposed UI2CODE pipeline first recalls UI elements with an object detector, then uses gradient cues to separate simple from complex regions and applies an SRGAN to restore foreground details in challenging backgrounds, achieving higher precision, recall, and localization than GrabCut and Deeplab, though it demands extensive multi‑scale training data.
AIGANObject Detection
0 likes · 4 min read

At QCon Beijing 2019, four Beike technology experts presented the practical use and challenges of machine learning for user profiling, deep‑learning‑based house‑quality scoring, intelligent customer‑service systems, and AI‑driven floor‑plan generation, summarizing the architecture, data pipelines, model evolution, and future improvement directions.
AIDeep LearningGAN
0 likes · 16 min read

This article explores multimodal representation learning by introducing a Bilinear Residual Layer that automatically fuses image and text features, demonstrates its superiority over traditional concatenation and FiLM methods on text‑guided image editing and fashion synthesis tasks, and reports state‑of‑the‑art results on several benchmark datasets.
GANbilinear residual layercross-modal attention
0 likes · 17 min read


21CTOMar 4, 2019 · Artificial Intelligence
This article explains why AI‑generated faces from StyleGAN are hard to distinguish, introduces an online game for testing realism, and provides practical visual cues—such as water spots, background errors, asymmetric glasses, hair artifacts, and teeth anomalies—to reliably identify fake images.
AI-generated imagesComputer VisionFace Detection
0 likes · 8 min read

This tutorial outlines a curriculum covering open‑source examples for enhancing image resolution using SRGAN, explains GAN‑based super‑resolution concepts, details network architectures and perceptual loss, and provides a simple MNIST training walkthrough with code links and resources.
GANMNISTSRGAN
0 likes · 7 min read

This article explains the design of a Cross‑Domain Embedding (CSDE) method that transfers Alipay user features to Taobao representations, details its learning and adaptive prediction stages, and shows experimental and online results demonstrating significant conversion‑rate improvements for new and inactive users.
Deep LearningGANRecommendation System
0 likes · 15 min read

This article explains the distinction between discriminative and generative models, outlines the challenges of limited labeled data, introduces semi‑supervised learning principles, and describes GAN‑based semi‑supervised classification algorithms with illustrative diagrams.
GANartificial intelligencegenerative models
0 likes · 3 min read


21CTOJul 22, 2018 · Artificial Intelligence
Researchers propose an unsupervised GAN-based image-to-image translation method that automatically dresses nude women in bikinis, preserving semantic content while removing sensitive parts, using unpaired datasets and Mask‑RCNN background removal, demonstrating impressive visual results without manual annotation.
Content moderationGANimage-to-image translation
0 likes · 10 min read

This article introduces Generative Adversarial Networks (GANs), explains their minimax formulation, value function, Jensen‑Shannon divergence, common variants, and practical training issues such as gradient saturation, while also previewing the next topic on Hidden Markov Models.
Deep LearningGANGenerative Adversarial Networks
0 likes · 11 min read


21CTOOct 9, 2017 · Artificial Intelligence
This article explains the challenges of image‑based porn detection, details the multi‑label classification approach of the Wukong system, and reveals the deep‑learning techniques—including CNN evolution, transfer learning, loss functions, adversarial training, and GAN‑based data augmentation—that enable over 99.5% accuracy with massive daily request volumes.
CNNGANadversarial training
0 likes · 18 min read

This article summarizes two SIGIR 2017 papers: one introduces a probabilistic latent‑class model for shopping‑pair push recommendations that improves e‑commerce click‑through rates by leveraging co‑purchase and view‑then‑purchase graphs, and the other presents IRGAN, a GAN‑based framework that unifies generative and discriminative information‑retrieval models, achieving state‑of‑the‑art results across web search, recommendation, and QA tasks.
GANe-commerceinformation retrieval
0 likes · 9 min read

Recent advances in image super-resolution leverage deep learning techniques such as convolutional neural networks, residual learning, perceptual loss, generative adversarial networks, and PixelCNN to reconstruct high-resolution details from low-resolution inputs, addressing challenges of scalability, training efficiency, and multi-scale upscaling.
CNNDeep LearningGAN
0 likes · 13 min read
