Author

Data Party THU

Official platform of Tsinghua Big Data Research Center, sharing the team's latest research, teaching updates, and big data news.

316

Articles

Likes

Views

Comments

Latest from Data Party THU

100 recent articles max

Data Party THU

Mar 8, 2026 · Artificial Intelligence

6 Practical Context‑Engineering Techniques to Tame RAG Hallucinations

This article explains why retrieval‑augmented generation (RAG) models often hallucinate, introduces the concept of context engineering, and details six practical techniques—including selective retrieval, context compression, hierarchical layout, dynamic query rewriting, memory management, and tool‑aware context—along with their trade‑offs and real‑world impact.

AIContext EngineeringLLM

0 likes · 23 min read

6 Practical Context‑Engineering Techniques to Tame RAG Hallucinations

Data Party THU

Mar 6, 2026 · Artificial Intelligence

How Small Can a Transformer Get? Inside the 121‑Parameter AdderBoard Challenge

This article chronicles the AdderBoard competition, detailing how researchers compressed a Transformer for 10‑digit addition down to just 121 parameters, the experimental rules, the contrasting hand‑coded and data‑driven approaches, and the insights gained about model minimalism and discoverability.

AdderBoardTransformermodel compression

0 likes · 13 min read

How Small Can a Transformer Get? Inside the 121‑Parameter AdderBoard Challenge

Data Party THU

Mar 4, 2026 · Artificial Intelligence

Can Hyperbolic Embeddings Boost Multi‑Step Visual Planning? Introducing GeoWorld

GeoWorld tackles the geometric neglect and multi‑step shortcomings of energy‑based predictive world models by mapping latent representations onto hyperbolic manifolds and applying a geometry‑aware reinforcement learning framework, achieving notable success‑rate gains on long‑horizon visual planning benchmarks.

World Modelsenergy-based modelsgeometric reinforcement learning

0 likes · 9 min read

Can Hyperbolic Embeddings Boost Multi‑Step Visual Planning? Introducing GeoWorld

Data Party THU

Mar 4, 2026 · Artificial Intelligence

Top 10 Must‑Install OpenClaw Skills for Powerful AI Agents

This article introduces OpenClaw’s versatile Skills system, highlights the ten most recommended Skills—including tavily-search, find‑skills, proactive‑agent, GitHub integration, and security tools—provides installation commands, and explains how each enhances AI agents’ capabilities for real‑time search, automation, and safe operation.

OpenClawSkills

0 likes · 5 min read

Data Party THU

Mar 2, 2026 · Artificial Intelligence

How ReLE Redefines Chinese LLM Evaluation and Reveals Capability Anisotropy

The ReLE framework introduces a dynamic, variance‑aware evaluation system that diagnoses capability anisotropy across 304 Chinese large language models, exposing ranking instability, commercial‑vs‑open‑source gaps, and format barriers while cutting evaluation cost by 70%.

AI assessmentCapability anisotropyChinese LLMs

0 likes · 9 min read

How ReLE Redefines Chinese LLM Evaluation and Reveals Capability Anisotropy

Data Party THU

Mar 1, 2026 · Artificial Intelligence

Unlocking Efficient LLM Fine‑Tuning: LoRA, QLoRA, and DoRA Compared

This article examines three parameter‑efficient fine‑tuning (PEFT) techniques—LoRA, QLoRA, and DoRA—explaining their core mechanisms, providing implementation code, benchmark results, memory and speed trade‑offs, and offering guidance on which method best fits different hardware and accuracy requirements.

DoRAFine-tuningLoRA

0 likes · 20 min read

Unlocking Efficient LLM Fine‑Tuning: LoRA, QLoRA, and DoRA Compared

Data Party THU

Feb 28, 2026 · Artificial Intelligence

How MIT’s Attention Matching Turns Linear Regression into Fast KV Compression

The article explains MIT’s Attention Matching technique that reformulates large‑model context compression as a linear regression problem, detailing its theoretical foundations, three‑step gradient‑free implementation, architectural adaptations, non‑uniform budgeting, and extensive evaluations showing orders‑of‑magnitude speed gains with minimal accuracy loss.

Attention MatchingKV compressionLinear regression

0 likes · 10 min read

How MIT’s Attention Matching Turns Linear Regression into Fast KV Compression

Data Party THU

Feb 27, 2026 · Artificial Intelligence

How “Vibe Coding” Is Redefining Software Development in 2026

Vibe coding, introduced by Andrej Karpathy, lets developers describe software functionality in natural language, letting large language models generate complete code, and the article reviews its concept, three leading 2026 tools (Cursor, Replit, Windsurf), a step‑by‑step workflow, advantages, drawbacks, and future trends.

AI codingLLMVibe Coding

0 likes · 10 min read

How “Vibe Coding” Is Redefining Software Development in 2026

Data Party THU

Feb 25, 2026 · Artificial Intelligence

Why Multimodal LLMs Miss Tiny Objects—and How to Fix It

This article analyzes why multimodal large language models often fail to detect small objects, identifies three core bottlenecks, and presents a four‑tiered optimization roadmap—from zero‑cost inference tricks to data augmentation, model fine‑tuning, and engineering safeguards—backed by three real‑world case studies and actionable guidelines.

data augmentationinference optimizationmodel fine-tuning

0 likes · 20 min read

Why Multimodal LLMs Miss Tiny Objects—and How to Fix It

Data Party THU

Feb 24, 2026 · Artificial Intelligence

Why Long Contexts Undermine LLM Reliability: Hidden Risks of Personalization and Shared Sessions

The article analyzes how expanding the context window of large language models creates scarce attention, introduces unreproducible personalization, mixes intents in shared accounts, and leads to performance degradation, making debugging, testing, and reliable production deployment increasingly difficult.

AI reliabilitycontext managementpersonalization

0 likes · 11 min read

Why Long Contexts Undermine LLM Reliability: Hidden Risks of Personalization and Shared Sessions