LangGraph Studio is a new visual IDE that simplifies the development, debugging, and interactive iteration of complex LLM‑based agent applications, offering features such as graph visualization, real‑time state inspection, code‑aware debugging, and seamless integration with LangSmith, with step‑by‑step guidance for desktop users.
This article explains what large language models (LLMs) are, how they are trained, their diverse applications across industries, the challenges they face, and provides a practical Python example using Replicate to run Meta's Llama 3‑70b‑instruct model.
This article details Meituan Waimai's feed advertising system, covering business characteristics, the evolution of estimation models, and practical implementations such as decision‑path modeling, ultra‑long/wide user modeling, full‑reconstruction techniques, and the integration of large language models for CTR prediction.
This article explains how large language models (LLMs) can launch data democratization by constructing a metric platform that combines LLM agents, semantic layers, NL2SQL/NL2API pipelines, warehouse‑internal and external semantics, and showcases SwiftAgent/SwiftMetrics innovations, real‑world case studies, and future directions.
This article explains the fundamentals of large language models, covering tokenization, probability prediction, Markov chain basics, training data limitations, context windows, and the transition to neural network architectures like Transformers, while providing Python examples and insights into model scaling and the illusion of intelligence.
This article explains how generative AI models work by demystifying tokens, tokenization with tools like tiktoken, simple Markov‑chain training, the limitations of small context windows, and how modern LLMs use neural networks, transformers and attention mechanisms to predict the next token.
This article reviews cutting‑edge research on integrating large language models with graph‑based recommendation systems, detailing four key strategies—LLM node embeddings, deep graph‑LLM fusion, model‑driven graph data training, and text‑modal enhancements—while analyzing representation learning, InfoNCE optimization, explainable recommendations, and extensive experimental validation.
This article introduces LangGraph, a Python library for creating stateful, multi‑agent LLM workflows, explains its loop, persistence, and human‑in‑the‑loop features, shows how to install it, and provides a complete code example that builds, runs, and reuses a searchable AI agent with thread‑level state saving.
The article examines DARPA's TRACTOR project, which aims to use large‑language‑model AI to automatically translate legacy C and C++ code into safe Rust, discusses the memory‑safety motivations, technical challenges, community reactions, and the broader push by US agencies to abandon unsafe languages.
The article presents a practical method for evaluating 1‑million‑token LLMs by inserting structured medal data into a classic Chinese novel, provides a full Python script for the test, shares results on GLM‑4‑long, and discusses training techniques and open‑source resources for long‑context models.
This article reviews four ACL 2024 papers authored by Meituan’s research team—covering training cost reduction, speculative decoding, code generation optimization, and instruction fine‑tuning—while also announcing a live sharing session at the conference.
The article breaks down multiple practical approaches for data preparation, training code handling, and experiment analysis in large‑language‑model fine‑tuning, showing how deeper engagement in each step can boost personal expertise even when final model performance appears similar.
MegaParse is an open‑source document parser that transforms PDFs, Word, PPT, Excel and CSV files into LLM‑friendly formats, preserving full information, boosting processing efficiency, and enabling deeper semantic analysis, with quick‑start installation steps and a roadmap for future features.
The article surveys the rapid emergence of large AI models and AIGC, explains core concepts like AI, AGI, and LLMs, details prompt‑engineering techniques such as chain‑of‑thought, outlines a seven‑layer AIGC stack, discusses technical and ethical challenges, and highlights future multimodal and industry‑specific applications.
This article analyses the shortcomings of current large language model training methods such as SFT, RLHF and DPO, explains why they incur high data and compute costs, and introduces Intuitive Fine‑Tuning (IFT) with temporal residual connections as a cheaper yet effective alternative that better aligns training objectives with real generation tasks.
This article introduces an AI‑driven Smart Report Assistant Bot that automatically converts uploaded Excel files into recommended charts, allows users to customize reports, and details the underlying workflow—including Excel parsing, LLM‑generated SQL, dynamic table creation, chart rendering with ECharts, and image‑merging plugins.
This article introduces STORM, a Stanford‑developed large‑language‑model‑based knowledge‑management platform that automates topic research, outline generation, citation‑rich article writing, and iterative refinement through perspective‑guided questioning and simulated conversations, dramatically improving technical investigation efficiency.
This article introduces STORM, a Stanford‑developed LLM‑based knowledge‑management system that automates topic research, outline creation, content generation with citations, and optimization through perspective‑guided questioning and simulated dialogue, dramatically improving technical investigation efficiency.
This guide walks through creating a full‑featured AI‑native API gateway using Higress, covering architecture setup, AI agent integration, observability, content security, token rate limiting, caching, retrieval‑augmented generation, prompt templates, and intelligent request/response transformation with concrete configuration examples.
This article explains the fundamentals of large language models, artificial general intelligence, AI-generated content, AI agents, retrieval‑augmented generation, knowledge bases, multimodal processing, fine‑tuning, alignment, tokens, vectors, and related tools, highlighting their capabilities, limitations, and practical considerations.
This report presents a continual pre‑training approach that significantly enhances Llama‑3 (8B)’s Chinese language proficiency and scientific reasoning by using a carefully mixed corpus of existing and synthetic data, detailing the bilingual adaptation and synthetic‑enhancement stages, data‑mixing and curriculum strategies, and demonstrating strong results across multilingual and scientific benchmarks without sacrificing original capabilities.
This article reviews the state‑of‑the‑art applications of large language models in recommendation systems, summarizing background knowledge, recent advances such as LLM4Rec, various tuning strategies, agent‑based approaches, open research problems, and future directions for generative recommendation.
This article surveys the design of autonomous large‑language‑model agents, detailing their modular architecture—including profiling, memory, planning, and execution—while also reviewing common profiling methods, memory structures, planning techniques, action strategies, and various learning approaches such as exemplar, human‑in‑the‑loop, and environment‑feedback training.
This article explains how to use the Shire language to create reliable coding agents by defining custom RAG workflows, leveraging IDE APIs, code verification functions, and vector‑based search, with detailed examples, configuration snippets, and a roadmap for future enhancements.
This article explains the challenges of deploying large language model (LLM) applications, outlines the need for end‑to‑end observability, and details Alibaba Cloud ARMS' LLM‑specific tracing, metrics, and Python agent solutions for monitoring, debugging, and performance optimization.
This article reviews the emerging field of large language model‑based AI agents, outlining their overall architecture, key challenges such as role‑playing, memory, planning, and multi‑agent collaboration, and discusses future research directions and practical examples in user behavior simulation and software development.
This article explores how large language model‑powered coding assistants boost developer productivity, shift the role of engineers toward higher‑level design and problem‑solving, and raise new responsibilities for code safety, performance, and ethical use in the evolving software development paradigm.
This article walks through the author’s experience creating a custom AI agent using LangChain and OpenAI APIs, explains the concepts of AI agents and the ReAct reasoning framework, provides step‑by‑step code, discusses required libraries and APIs, and shares practical tips and challenges encountered.
This talk explores how 360's advertising recommendation system leverages language models—from BERT to large‑scale LLMs—to improve user interest modeling, feature extraction, and conversion‑rate prediction, detailing practical challenges, engineering solutions, experimental results, and future research directions.
This article presents Task‑aware Decoding (TaD), a plug‑and‑play technique introduced by JD Tech and Tsinghua University and accepted at IJCAI 2024, which reduces intrinsic hallucinations in large language models by comparing pre‑ and post‑fine‑tuning outputs, and demonstrates its effectiveness combined with Retrieval‑Augmented Generation across various tasks.
This article provides a comprehensive overview of large language models (LLMs), covering their basic concepts, relationship with NLP, development history, parameter scaling, offline deployment, practical applications, prompt‑engineering frameworks, retrieval‑augmented generation, LangChain integration, agents, workflow orchestration, and future directions toward multimodal AI and AGI.
This article reviews JetBrains AI, an LLM‑powered assistant for JetBrains IDEs, exploring its code‑completion, code‑explanation, unit‑test generation, and refactoring capabilities through real C# examples and discussing its impact on developer workflows.
The talk by Huawei Noah's Ark Lab researcher Wang Yichao presents a comprehensive exploration of large language models (LLMs) for recommendation systems, covering background challenges, the KAR and Uni-CTR projects, experimental results, and future directions for open‑world, generative recommendation pipelines.
Mem0 is an open‑source memory‑management middleware for large language models that provides dynamic, context‑aware, and adaptive memory, outperforming traditional Retrieval‑Augmented Generation (RAG) and enabling personalized AI assistants, travel planners, and support agents with concrete Python APIs and examples.
This article walks through deploying the open‑source LLM serving tool Ollama and the Open‑WebUI interface on a Kubernetes cluster using Helm, covering GPU considerations, configuration files, service exposure, and model management with practical code examples.
This article surveys the rapid rise of AI agents powered by large language models, explaining their core perception‑planning‑action loop, memory architectures, tool‑use mechanisms, self‑reflection techniques, and real‑world case studies while highlighting current challenges and future prospects for autonomous intelligent systems.
Exploring Large Language Models, this article surveys their core concepts, evolution through Transformers, GPT and BERT, generation challenges, diverse applications such as QA, multimodal creation, summarization and retrieval‑augmented generation, prompt‑engineering frameworks and tools, LangChain‑based pipelines, AI‑driven agents, and future prospects toward domain‑specific use, multimodality, and AGI.
This article reviews the emerging field of large‑language‑model (LLM) based AI agents, outlining their overall architecture, core modules such as profiling, memory, planning and action, discussing current challenges, presenting concrete use‑cases, and highlighting promising research directions.
This article explains how Alibaba Cloud designed and deployed large‑language‑model agents for its service domain, covering background, ideal LLM deployment, the shift from explanation to problem solving, the agent framework, practical implementation, automation trade‑offs, training, evaluation, and real‑world impact.
Recent tests reveal that popular large language models—including GPT‑4o, Gemini Advanced, and Claude 3.5—often claim 9.11 is larger than 9.9 because their tokenizers split the numbers, but rephrasing, zero‑shot chain‑of‑thought prompts, or treating the values as floating‑point numbers can correct the mistake, a pattern also seen variably in Chinese models.
TaD, a task‑aware decoding technique jointly developed by JD.com and Tsinghua University and presented at IJCAI 2024, leverages differences between pre‑ and post‑fine‑tuned LLM outputs to construct knowledge vectors, significantly reducing hallucinations across various models, tasks, and data‑scarce scenarios, especially when combined with RAG.
This article reviews the hallucination problem in large language models, analyzes its data, training, and inference sources, and presents Task‑aware Decoding (TaD) and Retrieval‑Augmented Generation (RAG) as effective, plug‑and‑play solutions demonstrated through extensive experiments.
The article examines how AI-era large language model (LLM) applications impose new traffic, security, and scalability requirements on gateways, and explains how the Envoy‑based open‑source Higress gateway addresses these challenges with hot configuration updates, token‑based rate limiting, streaming support, and multi‑tenant capabilities.
This article explains how to use Alibaba Service Mesh (ASM) to register large language model (LLM) providers, configure LLMProvider and LLMRoute resources, and implement traffic routing, observability, and security for LLM services through step‑by‑step Kubernetes manifests and curl tests.
This article reviews the IJCAI 2024‑presented Task‑aware Decoding (TaD) technique, explains how it mitigates large‑language‑model hallucinations when combined with Retrieval‑augmented Generation, and details experimental results, practical deployments, and future research directions.
This guide explains how to deploy the open‑source LLM serving tool Ollama and the Open‑WebUI front‑end on a Kubernetes cluster using Helm charts, covering GPU configuration, persistent storage, service exposure, and model selection for large language models.
This guide compiles open‑source repositories, research papers, and practical tricks for training miniature large‑language models under 1 billion parameters, helping readers learn by reproducing models like nanoGPT, tinyLlama, Phi‑1.5, and more.
This article provides a comprehensive, step‑by‑step guide for developing large‑language‑model (LLM) applications, covering prompt design techniques, n‑shot and chain‑of‑thought strategies, retrieval‑augmented generation, structured I/O, workflow optimization, evaluation pipelines, operational best practices, and team organization to create reliable, scalable AI products.
This article presents a practical Python benchmark that measures token‑per‑second generation speed of various large language models—including GPT‑4o, glm‑4‑airx, and moonshot‑v1‑32k—by timing text generation on a Colab environment and summarizing the results in detailed tables and visual charts.
This guide explains how to enhance large language model outputs by constructing a GraphRAG pipeline that transforms relational data into a Neo4j knowledge graph, creates nodes and relationships via Python, and leverages LangChain's GraphCypherQAChain for accurate, context‑aware retrieval and generation.
The article presents Ant Group's research on integrating large language models into recommendation pipelines, covering background challenges, knowledge extraction, teacher‑model distillation, efficient deployment, experimental results, and future directions to improve accuracy and reduce bias.
This article provides a step‑by‑step guide for Java engineers on building a Retrieval‑Augmented Generation (RAG) application using the LangChain4j framework, covering RAG fundamentals, environment setup, Maven integration, document loading, splitting, embedding with OpenAI, vector store management with Chroma, and prompt‑based LLM interaction.
This article reviews the Kuaishou paper on adapting large language models for recommendation, detailing the LEARN framework's dual‑tower architecture, embedding generation, loss functions, and experimental results that address cold‑start and long‑tail challenges in modern recommender systems.
This tutorial walks you through creating a privacy‑preserving, locally hosted large language model chatbot using Ollama, Python 3, and ChromaDB, covering RAG fundamentals, GPU selection, environment setup, and full source code for a Flask‑based application.
This guide walks through the concepts, workflow, and practical steps for creating a high‑quality knowledge base in FastGPT using vector search, QA pair structuring, prompt engineering, and parameter tuning to improve LLM‑driven question answering.
genai is a Rust library that unifies the APIs of major large language models, offering a lightweight, native solution with simple chat‑focused examples, demonstrated through a dual‑model Rust program, and outlines future expansions such as additional model support, multimodal capabilities, and performance optimizations.
This article surveys how large language models reshape recommendation systems, presents a generative RS framework tailored for JD Alliance advertising, details material representation, model input, training and inference pipelines, and reports extensive offline and online experiments demonstrating its effectiveness on sparse user data.
This article explores the role of hyperparameters in large language models, explains each key hyperparameter, and guides readers through manual and automated tuning methods such as random search, grid search, and Bayesian optimization to achieve optimal model performance.
Vitron is a unified pixel‑level visual multimodal large language model that integrates image, video, and region encoders with a text‑centric strategy, delivering precise pixel‑wise perception and a comprehensive suite of vision tasks from understanding to generation and editing.
This article examines the practical challenges of using large language models in software development, including handling long contexts, cross‑file editing, bug‑fixing evaluation methods, and presents benchmark results from SWE‑Bench and its Lite subset to assess model capabilities.
This article provides a comprehensive overview of how large language models (LLMs) are applied in recommendation systems, covering two main paradigms—LLM+RS as a component and LLM as a standalone recommender—detailing their impact on pre‑training, fine‑tuning, prompting, and future research challenges.
This article surveys Retrieval‑Augmented Generation (RAG), analyzes the limitations of traditional vector‑based RAG, introduces Graph RAG that leverages knowledge graphs for more reliable context, proposes a universal RAG architecture compatible with vector, graph and full‑text indexes, and details its open‑source implementation, code components, testing, and future research directions.
This article surveys how large language models reshape recommendation systems, details a generative recommender framework for JD Alliance ads—including item representation, model input, training, and inference—presents extensive offline and online experiments, and discusses future optimization directions.
A recent Nature paper by MIT and collaborators argues that language serves mainly for communication rather than thought, presenting neuroscientific evidence that reasoning can occur without linguistic processing and sparking debate about the implications for large language models and artificial intelligence.
After more than a year of using LangChain for LLM‑driven applications, we found its heavy abstractions and inflexibility hindered production development, leading us to abandon it in 2024 and reconsider whether a dedicated AI framework is truly necessary.
This article walks through creating a LangGraph‑based multi‑agent application that automatically generates a movie scene, selects characters, simulates dialogue, and writes the script to a Word document using local Ollama‑powered LLMs and a custom workflow.
This guide walks through enabling KServe on Alibaba Cloud ASM, preparing the Llama‑2‑7B model with TensorRT‑LLM, creating the necessary Kubernetes resources, and deploying a serverless AI inference service that can be queried via a simple curl request.
The article describes how Google has integrated AI, particularly large language models, into its internal software development tools to improve developer productivity, outlines the challenges faced, shares lessons learned, and outlines future directions for AI‑driven engineering assistance.
This article demystifies large language models by comparing them to classic compression algorithms, explains how they compress massive data into compact parameters, explores their ability to learn abstract patterns, and provides practical insights into prompt engineering, sampling strategies, and multi‑step agent architectures for real‑world applications.
This article outlines a five‑week large‑model training curriculum, detailing weekly topics such as transformer fundamentals, encoder‑decoder architectures, self‑attention, LoRA fine‑tuning, and quantization, along with associated video lectures and PDF slide decks for developers.
This article presents a comprehensive guide to data engineering for research‑focused large language models, covering domain‑adaptive pre‑training, supervised fine‑tuning, retrieval‑augmented generation, dataset construction, data cleaning pipelines, token‑izer adaptation, and prompt engineering best practices to boost model performance in specialized tasks.
This guide walks through the complete workflow of creating an AI‑native 24‑point game using Baidu Cloud's AppBuilder, covering the three‑step methodology, Agent architecture, component design, custom workflow implementation, and practical tips for optimal model selection.
This article demonstrates how to protect large language model (LLM) requests in a cloud‑native environment by using Alibaba Cloud ASM service mesh and a custom Wasm plugin to dynamically inject API keys, enforce custom denial patterns, and optionally route requests through a private LLM for intelligent data‑leak detection.
This tutorial walks through installing LangChain, creating an LLM (via own GPU resources or third‑party APIs), designing parameterized prompt templates, implementing a custom output parser for structured results, and running a complete Python example that generates culturally specific names.
This article introduces LangChain, an open‑source framework for building LLM‑powered applications, outlines its key components such as prompts, chains, agents, and retrieval‑augmented generation, compares its advantages and drawbacks, and provides step‑by‑step instructions for setting up a Python development environment.
This article provides a comprehensive overview of the evolution of large language models, comparing foreign and Chinese offerings, explaining parameter significance, multimodal extensions, and detailing which models are available for commercial use.
This article introduces the Alibaba Cloud PAI platform for large language model (LLM) fine‑tuning, covering model‑training pipelines, performance‑cost trade‑offs, retrieval‑augmented generation, fine‑tuning methods such as full‑parameter, LoRA and QLoRA, model selection, data preparation, evaluation, and real‑world deployment examples.
This article explores how large language models can transform data workflows by using embeddings for semantic search, building private domain knowledge bases, generating SQL code from natural language with visualized results, and enhancing exploratory data analysis, outlining practical steps and benefits for enterprises.
This article analyzes Kuaishou’s May 2024 paper on LLM‑driven recommendation, detailing its dual‑tower architecture, contrastive learning of user and item embeddings, and a CVR‑auxiliary task that together improve cold‑start handling and boost both offline and online AUC metrics.
The paper introduces SymbCoT, a Symbolic Chain‑of‑Thought framework that translates natural‑language problems into symbolic form, plans, solves, and verifies reasoning steps, achieving significantly higher logical reasoning performance than traditional CoT methods across multiple benchmark datasets.
Evaluating Retrieval‑Augmented Generation (RAG) systems requires both component‑level and end‑to‑end metrics—such as context relevance, recall, answer relevance, and groundedness—and can be automated with tools like TruLens, RAGAS, LangSmith, and Langfuse, enabling systematic selection and optimization of LLM applications.
This guide explains what AI agents are, how they combine large language models with planning, memory, and tool use, and provides a step‑by‑step LangChain implementation—including environment setup, tool integration, and a runnable example that solves math and performs web searches.
This guide walks through creating a Retrieval‑Augmented Generation pipeline that harvests JD Cloud documentation, stores vector embeddings in ClickHouse, and serves queries via FastAPI, LangChain, a Qwen LLM, and a Gradio front‑end.
The report presents the open‑source Index‑1.9B family—base, pure, chat, and character variants—detailing benchmark results, pre‑training optimizations such as a normalized LM‑Head and deeper‑slim architectures, the importance of modest instruction data, alignment via SFT/DPO, role‑play enhancements with RAG, and acknowledges remaining safety and factual limitations.
This article explains Hugging Face's KV cache quantization technique, detailing how it reduces memory usage for long‑context LLM generation, the underlying quantization methods, implementation steps in 🤗 Transformers, benchmark results versus fp16, and the trade‑offs between speed, memory, and accuracy.
This article explains how AI-powered code generators create high‑quality code, outlines their capabilities such as language translation and documentation assistance, discusses safety and copyright concerns highlighted by research, and emphasizes that while popular, these tools should augment rather than replace developers.
In this talk, Alibaba Cloud's OpenSearch RAG team shares their year‑long journey of building retrieval‑augmented generation systems, covering data parsing, slicing, vectorization, hybrid retrieval, model fine‑tuning, performance optimizations, cost reduction, and future directions such as multimodal queries and agents.
This article surveys the impact of large language models on recommendation systems, outlines generative recommender architectures, discusses challenges of JD Union advertising, presents a semantic‑ID based solution with training and inference details, and reports offline and online experimental results.
This article surveys how large language models can reshape recommendation systems, describes the four-stage generative pipeline, details item representation techniques such as semantic IDs, presents a JD affiliate advertising use case with offline and online experiments, and outlines future optimization directions.
This article reviews how large language models are transforming recommendation systems, covering their fundamentals, recent LLM‑enabled methods for representation, learning and generalization, challenges such as scalability, bias and privacy, and future research directions including personalized prompts and robust model integration.
This article presents Xiaomi's work on LLM‑based Agent technology, covering its perception‑thinking‑action pipeline, technical framework, prompt management, executor and API platform, workflow, optimization strategies, evaluation metrics, and future directions for AI assistants.
The HandCat team designed an end‑to‑LLM pipeline that separates agent templates, tool protocols, and view layers, enabling LLM‑driven agents with memory, planning, and three tool types—general, selector, and interruptor—to safely manage sessions, handle errors, and balance granularity for performance within a commercial mobile app.
This tutorial shows Java engineers how to build a JUnit Rule that captures test failures, extracts the exception stack and source file, and automatically sends the information to OpenAI's GPT for analysis and code‑fix suggestions, complete with reusable data‑model classes and utility methods.
This article explores two advanced retrieval paradigms—Fusion Retrieval, which merges results from multiple retrievers using re‑ranking, and Recursive Retrieval, which builds hierarchical chunk‑to‑chunk or chunk‑to‑retriever links—to boost the quality and flexibility of Retrieval‑Augmented Generation pipelines.
This article introduces Reader, an open‑source Jina AI tool that converts any web URL into a format optimized for large language models, and llm.c, a minimalist C and CUDA project that demonstrates how to train a GPT‑2‑style LLM from scratch.
This article presents a practical approach to automatically generate software test cases by leveraging LangChain, PDF parsing, vector‑database retrieval, and large language models, comparing it with existing tools, detailing implementation steps, code examples, experimental results, and future improvement directions.
The article outlines LLM issues such as hallucination, outdated knowledge, and data privacy, then explains Retrieval‑Augmented Generation—detailing its data‑preparation and query‑time retrieval workflow, demonstrates a full LangChain implementation, and contrasts RAG with fine‑tuning as complementary strategies for up‑to‑date, grounded responses.
This tutorial walks through installing LangChain, creating an LLM with either self‑hosted or third‑party models, designing custom prompt templates, configuring output parsers for structured results, and running a complete Python example that generates culturally specific names using OpenAI's API.
This article explains the fundamentals and engineering practices of large language models (LLM), retrieval‑augmented generation (RAG) and AI agents, compares small and large embedding models, provides Python code for vector‑database RAG with Chroma, and discusses integration, use cases, and future challenges in AI development.
This article analyzes whether large‑scale web crawls, when meticulously filtered and deduplicated, can match or surpass the performance of high‑quality curated datasets in training large language models, covering dataset composition, processing pipelines, experimental results, scaling‑law implications, and future data‑efficiency strategies.
