Author

Alimama Tech

Official Alimama tech channel, showcasing all of Alimama's technical innovations.

255

Articles

Likes

1.9k

Views

Comments

Latest from Alimama Tech

100 recent articles max

Alimama Tech

Jul 11, 2025 · Artificial Intelligence

Inside Alibaba’s Taotian AI Team: From Werewolf Games to RecGPT

This article explores Alibaba’s Taotian technology team, revealing how its young engineers blend cutting‑edge AI research with playful projects like AI‑driven Werewolf games, develop large‑scale recommendation models such as RecGPT, and foster a culture of curiosity, rapid experimentation, and flat, tech‑first management.

AIGCR&D managementartificial intelligence

0 likes · 24 min read

Inside Alibaba’s Taotian AI Team: From Werewolf Games to RecGPT

Alimama Tech

Jul 9, 2025 · Artificial Intelligence

How to Make LLMs Recognize and Resolve Their Own Uncertainty

This article introduces ConfuseBench, a benchmark that classifies LLM uncertainty into document‑missing, ability‑limited, and ambiguous types, and presents methods—including retrieval, chain‑of‑thought, and clarification—to detect and actively resolve uncertainty, improving answer quality across diverse tasks.

ClarificationInquiryLLM

0 likes · 17 min read

How to Make LLMs Recognize and Resolve Their Own Uncertainty

Alimama Tech

Jun 25, 2025 · Artificial Intelligence

Introducing ROLL: A Scalable, User‑Friendly RL Framework for Large‑Scale LLM Training

ROLL is an open‑source reinforcement‑learning framework designed for large language model post‑training that combines multi‑task RL, agentic support, flexible algorithm configuration, elastic resource scheduling, and rich observability, delivering significant accuracy gains across benchmarks while remaining easy to use for researchers, product developers, and infrastructure engineers.

AI frameworkLarge Language ModelsOpen Source

0 likes · 11 min read

Introducing ROLL: A Scalable, User‑Friendly RL Framework for Large‑Scale LLM Training

Alimama Tech

May 14, 2025 · Artificial Intelligence

Deep Research‑Driven Risk Root‑Cause Analysis with Domain Graph Constraints for Large‑Scale Advertising Traffic

This article presents a large‑scale advertising risk‑control solution that combines deep‑research paradigms, domain‑graph constraints, and large language models to enable explainable, responsible, and high‑precision fraud detection, detailing system architecture, challenges, demo workflow, and future directions.

AIDeep Researchadvertising fraud

0 likes · 11 min read

Deep Research‑Driven Risk Root‑Cause Analysis with Domain Graph Constraints for Large‑Scale Advertising Traffic

Alimama Tech

May 12, 2025 · Artificial Intelligence

Universal Recommendation Model (URM): A General Large‑Model Recall System for Advertising

The article presents the Universal Recommendation Model (URM), a large‑language‑model‑based recall framework that integrates world knowledge and e‑commerce expertise through knowledge injection and prompt‑driven alignment, achieving significant offline recall gains and a 3.1% increase in ad consumption while meeting high‑QPS, low‑latency production constraints.

AdvertisingRecallhigh QPS

0 likes · 17 min read

Universal Recommendation Model (URM): A General Large‑Model Recall System for Advertising

Alimama Tech

Apr 23, 2025 · Artificial Intelligence

How AI Agents Outsmart Humans in the “Who Is Spy” Campus Challenge

The campus AI Agent competition showcased how large‑language‑model‑powered agents can reason, deceive, and collaborate in a social deduction game, revealing model performance trends, participant insights, and future directions for multi‑agent AI research.

AIAgent CompetitionLarge Language Models

0 likes · 6 min read

How AI Agents Outsmart Humans in the “Who Is Spy” Campus Challenge

Alimama Tech

Apr 23, 2025 · Artificial Intelligence

Distribution-aware Graph Prompt Tuning (DAGPrompT) for Heterophilic Graphs

Distribution‑aware Graph Prompt Tuning (DAGPrompT) tackles the pre‑training/downstream mismatch on heterophilic graphs by jointly applying low‑rank GLoRA adaptation and hop‑specific prompts that recast tasks as link‑prediction, yielding up to 4.79% accuracy gains and an average 2.43% improvement in few‑shot node classification.

Few-Shot LearningPretrainingPrompt Tuning

0 likes · 9 min read

Distribution-aware Graph Prompt Tuning (DAGPrompT) for Heterophilic Graphs

Alimama Tech

Apr 23, 2025 · Artificial Intelligence

Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning

The paper introduces an explainable LLM framework (ELLM‑rele) that uses chain‑of‑thought reasoning and a multi‑dimensional knowledge distillation pipeline to compress large‑model relevance judgments into lightweight student models, achieving superior offline relevance scores and online click‑through and conversion improvements in Taobao’s search advertising.

LLMchain-of-thoughtexplainability

0 likes · 17 min read

Explainable LLM-driven Multi-dimensional Distillation for E-Commerce Relevance Learning

Alimama Tech

Apr 17, 2025 · Artificial Intelligence

Ali Mama External Investment & Brand Algorithm Series

The Ali Mama External Investment & Brand Algorithm Series compiles recent KDD, NAACL, and SIGIR research on ad allocation, pacing, traffic estimation, and creative optimization, offering a comprehensive overview of cutting‑edge techniques that enhance advertising efficiency and brand performance across digital platforms.

Ali MamaKDDNAACL

0 likes · 1 min read

Alimama Tech

Apr 17, 2025 · Artificial Intelligence

PosterMaker: High-Quality Product Poster Generation with Accurate Text Rendering

PosterMaker leverages a ControlNet‑based TextRenderNet with character‑level visual features and a reward‑driven foreground‑extension detector to generate high‑quality product posters that accurately render Chinese text (over 90% sentence accuracy) while preserving product fidelity, and is already deployed in Alibaba’s AI creative tool.

Poster Generationcharacter-level featuresdiffusion models

0 likes · 18 min read

PosterMaker: High-Quality Product Poster Generation with Accurate Text Rendering