Tagged articles
reasoning
86 articles · Page 1 of 1
Researchers from City University of Hong Kong, Shenzhen University, USTC, and Oxford propose Seq-Drifting, a context-dependent drifting field method enabling one-step text generation for diffusion language models, achieving 11.2x speedup over 4-step baselines with Gen-PPL of 36.01 on 128-token generation, and validated on translation, summarization, and reasoning tasks.
Seq-Driftingcontext-dependent drifting fielddiffusion language models
0 likes · 11 min read

This analysis of NVIDIA's open-source Auto-Ontology reveals it generates probabilistic semantic catalogs from schemas and documents using LLMs, but lacks deterministic reasoning engines, RDF/OWL reasoners, or logical deduction — making it a useful accelerator for certified metric sharing, not a full ontology platform capable of automated inference.
Auto-OntologyLLMNVIDIA
0 likes · 10 min read

This guide details a 12-step process for creating executable ontologies using OntoL's Auto Workshop, covering prerequisites, iteration modes, prompt writing, canvas inspection, field and relationship management, data binding, loading, reasoning, self-testing, approval workflow, troubleshooting, and includes prompt templates, a complete example, and a checklist.
OntoLapproval workflowauto workshop
0 likes · 21 min read

OntoL's self-developed DSL simplifies ontology modeling and reasoning with simple syntax, multi-scenario reuse, strong system integration, unified semantic expression for graph computing, front-loaded constraints for reliability, inherent explainability for auditing, and full technical autonomy for sustainable evolution.
DSLOntoLexplainability
0 likes · 6 min read

This survey paper introduces a four-axis taxonomy for Retrieval-Augmented Generation (RAG) systems—efficiency, defense, interactivity, and reasoning—moving beyond simple vector database integration to analyze design trade-offs in evidence retrieval, security, user alignment, and multi-step reasoning.
DefenseEfficiencyInteractivity
0 likes · 38 min read

Contrasting DeepMind's GNoME discovering 2.2 million crystal structures with Apple's finding that LLMs fail simple math when irrelevant details are added, the article argues AI excels at generating candidates but lacks true reasoning, so humans must define value, set boundaries, and take responsibility while using AI for rule drafting, gap checking, translation, and simulation.
AIGNoMEHallucination
0 likes · 16 min read

AlphaGo co-author Thore Graepel argues that LLMs only scale intuitive pattern matching without a separate deliberative search mechanism, making them fundamentally different from AlphaGo's hybrid architecture and unreliable for high-stakes reasoning tasks.
AI ArchitectureAlphaGoLLM
0 likes · 13 min read

Renmin University researchers propose ME-Decoding, a method that selects candidate tokens by jointly optimizing probability confidence and semantic diversity via Mahalanobis distance, outperforming probability-only truncation on reasoning and open-ended generation across multiple models and temperatures with only 3% latency overhead.
EMNLP 2026LLM decodingMahalanobis distance
0 likes · 11 min read

The author details how LLMs failed in underwater battlefield simulations despite trying prompts, agents, RAG, and knowledge graphs, and explains why ontologies — formal, reasoning-capable world models — are essential for trustworthy AI decisions, illustrating with a concrete case and a six-step ontology engineering process.
AI decision-makingLLM limitationsOntoL
0 likes · 12 min read

GPT-6 Astra beats GPT-5.6 on automation, computer operation, reasoning, and safety benchmarks while costing 2.5× more, but loses on general knowledge and creative writing, positioning it as an agentic execution engine rather than a chat model.
AI benchmarksAPI pricingAgent Capabilities
0 likes · 10 min read

A 2020 MIT interview presentation by Giambattista Parascandolo, once dismissed as nonsense, outlined open‑ended reasoning, language‑driven planning, and self‑modifying agents—ideas that closely match the core concepts behind OpenAI’s recent o1 and o3 models.
AI researchOpenAIReinforcement Learning
0 likes · 8 min read

Choosing between OWL‑based ontologies and graph‑database inference engines fundamentally affects knowledge‑graph design: ontologies provide formal logical consistency and open‑world reasoning, while graph inference offers fast, flexible queries, with each suited to different constraints, scalability, and maintenance scenarios.
Neo4jOWLTigerGraph
0 likes · 9 min read

The article reviews recent AI agent research, exposing three structural paradoxes—memory, reasoning, and evolution—each illustrated with concrete systems, benchmarks, and safety studies, and argues that only coordinated progress across all three dimensions can yield truly mature, self‑improving agents.
AI AgentsHarness EngineeringRecursive Self-Improvement
0 likes · 16 min read

At ACL 2026 in San Diego, Bilibili’s tech team introduced the FATE series—SABER, which reduces overthinking in LLMs with a token‑budgeted switchable training, and CASTER, a community‑aware evaluation system built on Social‑CoT and the MEDEA framework that outperforms GPT‑5.2 and Claude‑4.5‑Opus on the new CASTER‑Bench, while also promoting the B‑UP talent recruitment program.
BenchmarkLLMReinforcement Learning
0 likes · 10 min read


Qborfy AIJun 25, 2026 · Artificial Intelligence
This article explains how inference‑oriented LLMs differ from standard models, details the three key reasoning‑control parameters (reasoning_effort, thinking, max_completion_tokens), compares their usage across OpenAI o1, Anthropic Claude, DeepSeek‑R1 and Kimi, and provides practical guidance on selecting values, managing costs, and deciding when to employ reasoning models.
API parametersLLMcost control
0 likes · 18 min read

The article explains how prediction, reasoning, and inference differ, why a pure prediction model leaves the decision chain broken, and how a dynamic ontology‑driven feature framework—temporal, functional, and relational features—creates explainable, verifiable, and iterative decision loops.
decision makingdynamic ontologyfeature engineering
0 likes · 9 min read

The article demonstrates how an ontology can serve as a business‑semantic layer for enterprise AI agents, covering multi‑relationship propagation, schema‑to‑concept mapping, cross‑system customer views, and a unified semantic query engine, while also discussing practical limits and rollout advice.
AI Agentsenterprise AIknowledge graph
0 likes · 11 min read

This article explains how OWL extends RDF to model domain ontologies, outlines its core concepts, sub‑languages, key reasoning features, tooling, typical applications such as knowledge graphs and AI, and discusses both its advantages and practical challenges.
LLMOWLRDF
0 likes · 17 min read

OneReason introduces a systematic reasoning capability into industrial recommendation models through multi‑stage pre‑training, chain‑of‑thought fine‑tuning, and reinforcement learning, achieving significant gains in click‑through, revenue, and cross‑domain recommendation performance while preserving the underlying language abilities of the base model.
PretrainingReinforcement Learningchain-of-thought
0 likes · 29 min read

The OneReason report details how Kuaishou’s recommendation team injects reasoning into large‑scale recommender models through a four‑level pre‑training pipeline, chain‑of‑thought (CoT) fine‑tuning, and specialized reinforcement learning, achieving significant offline gains and a 10.33% exposure lift in a live A/B test.
CoTLLMPretraining
0 likes · 31 min read

The paper introduces PreRL, which removes the input condition to directly optimize the reasoning trajectory (P(y)) of large language models, and combines it with standard RL in Dual Space RL (DSRL), achieving consistent gains on math and out‑of‑distribution benchmarks, faster training, and richer reasoning behaviors.
DSRLMath BenchmarksPreRL
0 likes · 11 min read

Xiaomi’s AI team extracted 5 billion video frames to create the world’s largest open‑source GUI dataset, demonstrated that a 3 B‑parameter model can reduce inference tokens by 71% while surpassing larger models, and presented a suite of ICML 2026 papers covering data scaling, benchmarking, reasoning, multimodal perception, and training stability for GUI agents and other AI tasks.
DatasetGUI Agentbenchmarking
0 likes · 21 min read

The paper introduces TEMPO, a test‑time training framework inspired by the Expectation‑Maximization algorithm, which alternates policy optimization (M‑step) with Critic calibration (E‑step) to prevent reward‑signal drift, and demonstrates on Qwen3 and OLMO3 models that it continuously improves reasoning performance and maintains output diversity beyond the saturation point of existing TTT methods.
EM algorithmReinforcement LearningTest-Time Training
0 likes · 14 min read

The SHAPE framework (Stage‑aware Hierarchical Advantage via Potential Estimation) adds a milestone‑based “reasoning tax” to large language model inference, providing step‑wise correctness signals and penalizing verbosity, which yields an average 3% accuracy gain and a 30% reduction in token consumption across multiple math‑reasoning benchmarks.
ACL 2026LLMMathematical Reasoning
0 likes · 10 min read

The article explains ReAct—a Reason‑Act framework for large language model agents that observes, reasons, takes actions via tools, receives feedback, and iterates—highlighting its distinction from plain QA, its step‑by‑step workflow, practical importance, and a weather‑query example.
AI workflowLLM agentsReact
0 likes · 5 min read

This article analyzes the fundamental limitations of traditional vector‑based Retrieval‑Augmented Generation, introduces Vectify AI’s reasoning‑driven PageIndex framework, and explains how hierarchical, non‑vector indexing enables more accurate, context‑aware document retrieval for complex, domain‑specific texts.
AIDocument RetrievalLLM
0 likes · 15 min read

Recent studies reveal that in Agentic RL, jointly training reasoning and tool-use on shared parameters creates a persistent negative interaction, with gradients nearly orthogonal, limiting performance; a disentangled tuning approach (DART) using separate LoRA adapters isolates the two abilities and restores gains across benchmarks.
DartGradient InterferenceLoRA
0 likes · 12 min read

This article walks you through why enterprise AI agents need a semantic ontology, explains TBox and ABox concepts, outlines a general modeling workflow, introduces RDF/OWL standards and tools like Protégé and reasoners, and provides a hands‑on example—including Python code with Owlready2—to create and test a business ontology for order‑expedition rules.
OWLRDFknowledge graph
0 likes · 18 min read

This article outlines six concrete criteria—relationship‑centric data, reproducible reasoning, evolving schemas, multi‑hop queries, explainable decisions, and cross‑system data integration—to help engineers decide whether a knowledge graph is the right solution or if a relational database will suffice.
AI EngineeringGraph Databasesdata integration
0 likes · 15 min read

At the start of 2026, self‑distillation dominates the most cited LLM papers, offering a teacher‑free way for large models to continually acquire new knowledge while preserving existing capabilities.
Continual LearningReinforcement Learninglarge language models
0 likes · 9 min read

Google’s recent study of over 8,000 reasoning tasks shows that advanced LLMs like DeepSeek‑R1 spontaneously develop multiple internal “expert” personas that debate, and that activating a discovered “social switch” dramatically raises accuracy, revealing that engineered conflict can enhance AI reasoning.
AI debateFeature ControlLLM
0 likes · 8 min read

This article examines why most enterprise AI agents stumble—due to hallucinations, semantic mismatches, and lack of explainability—then introduces ontology as a semantic layer that structures business concepts, rules, and constraints to enable reliable reasoning, centralized rule management, and transparent AI behavior.
agententerprise-aiknowledge-graph
0 likes · 17 min read

At EMNLP 2025, the BUPT NIRC team presented a paper that introduces the ARR metric to quantitatively separate latent reasoning from factual shortcuts in LLMs, using Logit Lens and Attention Knockout to reveal distinct internal pathways and shares their conference experience.
ARR metricAttention KnockoutEMNLP2025
0 likes · 6 min read

This article explains the ReAct framework, which augments large language models with explicit reasoning, tool use, and observation loops to overcome hallucinations, improve transparency, and enable dynamic interaction with external environments across diverse tasks.
AI AgentsReactTool Use
0 likes · 31 min read

The article breaks down how AI product managers can transform a raw base model into a market‑ready, high‑impact product by applying supervised fine‑tuning, tool‑use routing, RLHF alignment, and chain‑of‑thought reasoning, while highlighting trade‑offs, cost shifts, and evaluation metrics.
RLHFSFTTool Use
0 likes · 13 min read

GLM-4.7 scores 68 points to rank sixth worldwide and first among open‑source models, surpassing Claude 4.5 Sonnet, with strong reasoning performance, fast generation speed, but higher cost and weaker code‑generation and math abilities compared to rivals.
BenchmarkGLM-4.7code generation
0 likes · 7 min read

DeepAgent, a new end‑to‑end reasoning agent, integrates autonomous thinking, dynamic tool search, and execution to handle over 16,000 APIs, embodied tasks, and research assistance, achieving state‑of‑the‑art performance on benchmarks like TMDB, ToolBench, ALFWorld, WebShop, and GAIA.
Reinforcement Learninglarge language modelmemory management
0 likes · 15 min read

This article explains how Chain‑of‑Thought prompting and its variants—zero‑shot CoT, Thread of Thought, Tabular CoT, Analogical Prompting, and Step‑back Prompting—enable large language models to perform multi‑step reasoning by breaking problems into intermediate steps, with practical prompts, examples, and implementation details.
Zero-Shot Learningchain-of-thoughtreasoning
0 likes · 12 min read

Interviewers increasingly ask why modern reward models must go beyond scalar scores to incorporate reasoning, and this article explains the limitations of traditional scalar reward models, the benefits of the RM‑R1 framework, and how reasoning‑based rewards improve alignment, stability, and task performance in large language model training.
AI AlignmentLLMRLHF
0 likes · 11 min read

VitaBench, a new open‑source benchmark from Meituan’s LongCat team, evaluates LLM‑driven agents across three realistic life‑service scenarios—food ordering, restaurant dining, and travel planning—using 66 tools and quantifying reasoning, tool, and interaction complexities, exposing a mere 30% success rate on complex cross‑scene tasks.
AIBenchmarkInteraction
0 likes · 14 min read

The article explains how massive language models achieve seemingly intelligent behavior through emergence at a critical scale, hierarchical planning, attention-driven global coherence, multimodal understanding, and progressive training techniques that turn simple token prediction into sophisticated reasoning and creativity.
Attention MechanismEmergencelarge language models
0 likes · 15 min read

Meituan’s LongCat team unveiled LongCat-Flash-Thinking, an open‑source large language model that combines deep logical reasoning with tool‑calling capabilities, achieving state‑of‑the‑art performance across logic, mathematics, code, and agentic tasks, and introducing novel training frameworks such as domain‑parallel RL and DORA.
AIBenchmarkTool Use
0 likes · 7 min read

This article surveys recent advances in applying reinforcement learning to large reasoning models, outlining the historical background, key breakthroughs like OpenAI o1 and DeepSeek‑R1, current challenges in reward design and scalability, and future research directions toward more capable AI systems.
AI researchRLHFSurvey
0 likes · 9 min read

This article examines how large language model reasoning has progressed from linear Chain‑of‑Thought prompting to parallel Tree‑of‑Thought and flexible Graph‑of‑Thought approaches, highlighting each method’s mechanism, strengths, limitations, computational costs, and the broader shift toward cognitive‑centric AI research.
AI researchGraph-of-ThoughtTree of Thought
0 likes · 7 min read

Atom-Searcher introduces an atomic‑thought reinforcement‑learning framework that decomposes complex reasoning into fine‑grained units, uses a Reasoning Reward Model to assign step‑wise rewards, dynamically balances process and result incentives, and achieves state‑of‑the‑art performance on multiple LLM benchmarks.
Agentic ResearchAtomic ThoughtLLM
0 likes · 12 min read

The article analyzes Denny Zhou's Stanford CS25 lecture on large language model reasoning, explaining how intermediate token generation, chain‑of‑thought prompting, self‑consistency, reinforcement‑learning fine‑tuning, and answer aggregation together unlock powerful reasoning capabilities beyond traditional greedy decoding.
AI researchLLMSelf-Consistency
0 likes · 17 min read

This article distills Denny Zhou’s Stanford CS25 lecture, explaining how large language models achieve reasoning through intermediate token generation, chain‑of‑thought prompting, self‑consistency, reinforcement‑learning fine‑tuning, and answer aggregation, while highlighting theoretical foundations and practical breakthroughs.
LLMReinforcement Learningchain-of-thought
0 likes · 18 min read

The article explores how Von Neumann’s insights on the brain‑computer analogy illuminate modern large‑language‑model prompt engineering, comparing logical reasoning chains, memory mechanisms, and DSL‑driven computation to improve accuracy, reduce hallucinations, and balance reasoning depth with precise calculation.
DSLRAGcomputation
0 likes · 14 min read


Amap TechJul 24, 2025 · Artificial Intelligence
The paper introduces FingER, an entity-level evaluation framework and the FingER-Instruct-60k dataset for assessing AI-generated video quality with fine-grained reasoning, and demonstrates state-of-the-art zero-shot performance on multiple benchmarks using novel training strategies.
AI-generated videoDatasetfine-grained evaluation
0 likes · 9 min read

Since GPT‑4’s 2023 debut, large language models have shifted from sheer scale to efficiency‑driven designs, advanced reasoning with chain‑of‑thought, and agentic tool use, as illustrated by MoE, MLA, and new attention mechanisms, reshaping benchmarks, commercial strategies, and the future of AI.
Agentic AIEfficiencyLLM
0 likes · 24 min read

This article analyzes the rapid evolution of large language models from the GPT‑4 era through efficiency‑focused sparsity and attention innovations, to inference‑time reasoning and tool‑using agents, highlighting key architectures, benchmark breakthroughs, competitive strategies, and emerging research directions toward embodied AI.
Agentic AIEfficiencyLLM
0 likes · 24 min read

This article traces the evolution of large language models from the GPT‑4 era through 2024‑2025, highlighting the shift from pure scaling to efficiency‑focused architectures, the rise of reasoning‑centric "thinking" models, and the emergence of agentic capabilities that enable tools and real‑world interaction.
AgentsLLMTransformer
0 likes · 27 min read

This roundup presents recent AI breakthroughs, including Robo‑DM’s efficient robot dataset management, DeepKD’s decoupled knowledge‑distillation trainer, a novel informed white‑box attack exposing weaknesses in LLM alignment defenses, the RePPL hallucination detector, Self‑GIVE’s associative reasoning framework, and LLM‑driven RL ensemble methods.
AIknowledge distillationreasoning
0 likes · 15 min read

This paper introduces the Thought Leap Bridge task and the CoT‑Bridge model, which detect and fill missing intermediate steps in chain‑of‑thought reasoning, dramatically improving large language model performance on mathematical and logical benchmarks and enhancing downstream distillation and reinforcement‑learning pipelines.
Chain-of-ThoughtCoT-BridgeLLM
0 likes · 8 min read

Researchers from NUS, Tsinghua and Salesforce AI Research introduce a meta‑capability alignment framework that integrates deductive, inductive and abductive reasoning via a psychology‑based triple, automatically generates and validates training data, and demonstrates over 10% accuracy gains on math, coding and scientific benchmarks for 7B and 32B models.
Meta‑Capability AlignmentModel Traininglarge language models
0 likes · 8 min read

DeepSeek‑R1‑0528, the latest open‑source 660B LLM, dramatically improves coding, reasoning, and long‑context abilities, matching or surpassing top models like Gemini 2.5 Pro and Claude 4 in benchmarks and real‑world tests, while offering faster, more stable, and fully executable outputs.
AI benchmarkingDeepSeekcode generation
0 likes · 7 min read

This article introduces the OmniThought dataset, which annotates over two million chain‑of‑thought reasoning steps with Reasoning Verbosity and Cognitive Difficulty scores, and explains how these metrics guide the training of DistilQwen‑ThoughtX models that adapt chain length to task difficulty, achieving superior performance compared to existing distilled LLMs.
CoTDatasetLLM
0 likes · 16 min read

This guide systematically explores deductive, inductive, analogical, and causal reasoning, outlines common logical fallacies, presents argument structures and evaluation criteria, and introduces the Toulmin model, offering readers practical tools to strengthen critical thinking and construct persuasive, well‑grounded arguments.
Fallaciesargumentationcritical thinking
0 likes · 6 min read


DevOpsMay 18, 2025 · Artificial Intelligence
Although large language models have enabled AI agents that mimic human digital interactions, their commercial accuracy remains far below production standards, prompting the industry to pivot toward agentic workflows and data synthesis, which promise more reliable task automation, reasoning, and observable, auditable processes for knowledge work.
Agentic workflowsData Synthesisknowledge work
0 likes · 6 min read

The article introduces Chain‑of‑Recursive‑Thoughts (CoRT), explains how recursive self‑evaluation enhances large language model reasoning, outlines its workflow, shares GitHub resources, compares it with existing CoT methods, and reports experimental results using Mistral 3.1 24B.
AIChain-of-Recursive-ThoughtsCoRT
0 likes · 6 min read

This article analyzes whether reinforcement learning enhances large language model reasoning, compares findings from DeepSeek-Math, a Tsinghua‑Shanghai Jiao‑Tong paper, and Qwen3, and outlines practical training pipelines—including Seed‑Thinking‑v1.5, DeepSeek‑R1, Kimi‑K1.5, and Qwen3—that aim to endow LLMs with robust reasoning capabilities.
LLMModel TrainingReinforcement Learning
0 likes · 12 min read

Recent empirical research by Tsinghua’s LeapLab and Shanghai Jiao Tong University reveals that reinforcement‑learning‑based fine‑tuning (RLVR) improves sampling efficiency but does not extend the fundamental reasoning abilities of large language models beyond their base capabilities, as demonstrated across mathematics, code, and visual reasoning benchmarks.
AI researchRLVRReinforcement Learning
0 likes · 12 min read

UC Berkeley researchers found that forcing large language models to skip explicit reasoning—using a “NoThinking” mode—can achieve comparable or better accuracy with significantly fewer tokens, especially under token budget constraints, across math, coding, and theorem‑proving benchmarks.
NoThinkingreasoningtoken efficiency
0 likes · 7 min read


AntTechApr 10, 2025 · Artificial Intelligence
At the ICLR 2025 live session in Singapore, Ant Group showcased four cutting‑edge papers—CodePlan, Animate‑X, Group Position Embedding, and OmniKV—demonstrating advances in large‑language‑model reasoning, universal character animation, layout‑aware document understanding, and efficient long‑context inference.
AI researchDocument Understandinglarge language models
0 likes · 6 min read


JavaEdgeMar 24, 2025 · Artificial Intelligence
This article examines the fundamental reasoning limitations of large language models, illustrates real‑world failure cases, and outlines current research directions such as better datasets, chain‑of‑thought prompting, external verification, and specialized solvers to improve their logical capabilities.
AILLMchain-of-thought
0 likes · 8 min read

The AI SDK 4.2 release introduces powerful new features such as step‑by‑step reasoning support, a Model Context Protocol (MCP) client for tool integration, useChat message components, multimodal image generation, standardized URL sources, OpenAI Responses API support, Svelte 5 compatibility, and numerous middleware and provider enhancements, all illustrated with practical JavaScript/TypeScript examples.
AI SDKImage GenerationJavaScript
0 likes · 19 min read

DeepSeek’s R1 inference mode replaces linear chain‑of‑thought with a transparent, multi‑path tree reasoning system, offering layered analysis, intent understanding, memory management, emotion detection, and hallucination mitigation, illustrated through a practical example of buying authentic cigarettes and detailed technical breakdowns.
HallucinationMemoryartificial intelligence
0 likes · 16 min read

This article provides a comprehensive overview of AI Agents, tracing their evolution from traditional chatbots to LLM‑driven agents, explaining core components such as perception, reasoning, action, knowledge bases, learning and communication interfaces, and discussing practical use cases, interaction cycles, and future prospects.
AI AgentsRetrieval-Augmented GenerationTool Use
0 likes · 15 min read


ArchitectMar 3, 2025 · Artificial Intelligence
This article examines how to build and improve reasoning‑capable large language models, explains the definition and use‑cases of reasoning models, details DeepSeek‑R1’s training pipeline, compares four key enhancement methods—including inference‑time scaling, pure RL, SFT + RL, and distillation—and offers budget‑friendly advice.
AI researchDeepSeekInference Scaling
0 likes · 27 min read

Elon Musk’s xAI introduced the Grok 3 family—trained on roughly 200,000 GPUs and offered in standard, mini and Reasoning versions—that claims top‑slot performance on math, science and coding benchmarks, outpacing Google Gemini, DeepSeek V3, Claude and OpenAI GPT‑4o, while pricing starts at $30 per month and drawing both praise for its speed and criticism for lingering hallucinations and ethical sensitivities.
AIBenchmarkDeepSearch
0 likes · 16 min read

DeepSeek AI’s new open‑source model DeepSeek‑R1 leverages a novel Group‑Related Policy Optimization (GRPO) reinforcement‑learning framework and multi‑stage training to dramatically boost complex reasoning performance, achieving AIME 2024 Pass@1 scores comparable to OpenAI’s o1 model.
AIDeepSeekGRPO
0 likes · 4 min read

The article explains how tokenization mismatches and biased next‑token prediction cause LLMs to miscount letters in “Strawberry” and incorrectly compare 9.9 versus 9.11, and shows that step‑by‑step Chain‑of‑Thought prompting with reason‑first output dramatically improves accuracy.
AIChain-of-ThoughtLLM
0 likes · 13 min read

The article analyses the rapid growth of Retrieval‑Augmented Generation (RAG) in enterprises, explains why self‑built RAG solutions often collapse under cost and maintenance pressures, and demonstrates how Tencent LeXiang AI Assistant addresses these issues through a robust knowledge‑management core, extensive industry experience, scalable resources, and advanced multimodal capabilities.
AI assistantRAGenterprise AI
0 likes · 16 min read

The article introduces Marco‑o1, an open‑source LLM that enhances complex reasoning by fine‑tuning on Chain‑of‑Thought data, integrating Monte‑Carlo Tree Search, introducing mini‑step actions and a reflection mechanism, and evaluates its performance on multilingual math and translation benchmarks.
LLMMonte Carlo Tree Searchartificial intelligence
0 likes · 15 min read


CSS MagicSep 14, 2024 · Artificial Intelligence
The article examines OpenAI’s newly released o1 model, highlighting its superior performance in complex reasoning tasks such as math, programming, and science, and explains how model‑level chain‑of‑thought optimization and product‑level UI design give it an edge over competitors like Claude.
AI-evaluationChatGPTOpenAI
0 likes · 8 min read

OpenAI’s new o1 series, including o1‑preview and o1‑mini, leverages reinforcement‑learning‑based chain‑of‑thought reasoning to achieve superior performance on academic exams, coding contests, and safety benchmarks, offering faster, cost‑effective options while advancing AI alignment and human‑preference evaluation.
AI safetyBenchmarkOpenAI
0 likes · 15 min read

This article presents a comprehensive exploration of knowledge‑graph‑based modeling, neural‑symbolic multi‑hop reasoning, and large‑model‑driven question answering applied to precise medication decision‑making in hypertension, detailing system architecture, experimental evaluations, real‑world deployments, and future research directions.
hypertensionknowledge graphlarge language model
0 likes · 26 min read

This article explains ten common types of logical reasoning used in daily life, describing each concept, practical scenarios, and visual examples to help readers analyze information, make better decisions, and communicate more persuasively.
critical thinkingdecision makinglogic
0 likes · 8 min read

Alibaba Cloud’s AI platform PAI had four papers accepted at EMNLP 2023, presenting advances in automatic prompt engineering for text‑to‑image, domain‑specific knowledge‑enhanced language models, cognitive‑tree reasoning with small LLMs, and cross‑lingual machine reading comprehension, all demonstrating cutting‑edge AI research and product integration.
artificial intelligenceknowledge graphsprompt engineering
0 likes · 9 min read

This article reviews recent research on augmenting language and multimodal models with external knowledge sources and tool‑calling mechanisms, covering three systems—OREO‑LM for knowledge‑graph reasoning, REVEAL for multi‑source visual‑language pretraining, and AVIS for dynamic tool selection—and their experimental results and implications.
Language Modelknowledge graphmultimodal
0 likes · 28 min read

This article reviews recent research on augmenting language and vision models by incorporating external knowledge sources such as knowledge graphs, multi‑source retrieval, and dynamic tool‑calling frameworks, presenting three systems—OREO‑LM, REVEAL, and AVIS—and their experimental results.
AI researchLanguage Modelmultimodal
0 likes · 27 min read

The survey shows that once language models exceed roughly ten billion parameters they spontaneously acquire two complementary abilities—step‑by‑step reasoning, often elicited by chain‑of‑thought prompts or scratch‑pad training, and the capacity to invoke external tools such as search engines, calculators, or robots—enabling them to retrieve up‑to‑date information, perform complex computations, and act in the world, thereby advancing toward general artificial intelligence.
AIRetrieval AugmentationTool Use
0 likes · 20 min read

This article reviews recent advances in knowledge graph reasoning, introduces the RNNLogic framework that jointly learns a rule‑generating LSTM and a stochastic logic programming predictor, and demonstrates its competitive performance and interpretability on benchmark datasets while outlining future neural‑symbolic directions.
AIRNNLogicknowledge graph
0 likes · 10 min read

This teaching plan outlines a comprehensive Knowledge Graph course covering fundamentals, representation, storage, extraction, reasoning, fusion, question answering, graph algorithms, and emerging technologies across nine detailed chapters, including language integration, ontology matching, and multimodal extensions.
Graph Algorithmsartificial intelligencecourse syllabus
0 likes · 4 min read

This article surveys knowledge‑graph reasoning, explaining deductive and inductive reasoning fundamentals, description‑logic and logic‑programming approaches, and modern embedding techniques such as TransE, TransH, TransR and TransD, while highlighting their theoretical bases, practical implementations and recent research progress.
AIEmbeddingdeductive reasoning
0 likes · 13 min read

The talk reviews the history and concepts of knowledge graphs, explains logical and statistical reasoning methods—including rule‑based and representation‑learning approaches—and demonstrates how these techniques can be applied to build an intelligent robot that assists students in solving high‑school exam problems.
Representation Learningexam robotknowledge graph
0 likes · 5 min read
