Data Party THU
Author

Data Party THU

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

507
Articles
0
Likes
2.4k
Views
0
Comments
Recent Articles

Latest from Data Party THU

100 recent articles max
Data Party THU
Data Party THU
Jul 26, 2026 · Artificial Intelligence

VideoChat3: Open-Source Full-Stack Video Understanding Model Linking Perception, Understanding, and Interaction

VideoChat3 is a 4‑billion‑parameter multimodal large model that unifies short‑video, long‑video, and streaming video understanding through a native spatiotemporal encoder, adaptive resolution budgeting, and four‑stage training, achieving competitive accuracy while dramatically reducing visual token count and inference cost.

Adaptive ResolutionI3D-ViTMultimodal LLM
0 likes · 15 min read
VideoChat3: Open-Source Full-Stack Video Understanding Model Linking Perception, Understanding, and Interaction
Data Party THU
Data Party THU
Jul 26, 2026 · Artificial Intelligence

Understanding VLA Safety: A Visual Overview and Design Guidelines for Robot Security

The article reviews the Vision‑Language‑Action (VLA) safety landscape, classifies attacks and defenses across training and inference phases, highlights the multimodal attack surface, real‑time constraints, and simulation‑to‑reality gaps, and proposes a fast‑slow dual‑loop defense architecture for safe embodied AI.

AI safetyMultimodal AttackSimulation-to-Reality
0 likes · 9 min read
Understanding VLA Safety: A Visual Overview and Design Guidelines for Robot Security
Data Party THU
Data Party THU
Jul 25, 2026 · Artificial Intelligence

Generative Communications: A Controllable Generation Paradigm for 6G

This article reviews the emerging concept of Generative Communications (GenCom) for 6G, explaining how the communication goal shifts from raw data replication to controlled generation using minimal semantic cues, large‑model priors, and shared knowledge bases, and discusses its architecture, key technologies, applications, and open research challenges.

6GAI-driven NetworksControlled Generation
0 likes · 15 min read
Generative Communications: A Controllable Generation Paradigm for 6G
Data Party THU
Data Party THU
Jul 25, 2026 · Artificial Intelligence

Kimi K3 vs GPT‑5.6 Sol: A Full‑Scale Comparative Evaluation

The article presents a detailed head‑to‑head assessment of the open‑source Kimi K3 model and the closed‑source GPT‑5.6 Sol, measuring their ability to generate playable 3D games, handle full‑stack development tasks, and comparing performance, token efficiency, and engineering completeness.

3D game generationAI model comparisonGPT-5.6 Sol
0 likes · 13 min read
Kimi K3 vs GPT‑5.6 Sol: A Full‑Scale Comparative Evaluation
Data Party THU
Data Party THU
Jul 24, 2026 · Artificial Intelligence

Create an IMDB Sentiment Classifier with Scikit‑LLM & Groq in 500 Lines

This tutorial walks through building a complete sentiment‑analysis pipeline for the IMDB movie‑review dataset using Scikit‑LLM and Groq, from environment setup and data sampling to custom text cleaning, zero‑shot LLM classification, model fitting, prediction, and evaluation with a 95% accuracy report.

GroqIMDB datasetPython
0 likes · 8 min read
Create an IMDB Sentiment Classifier with Scikit‑LLM & Groq in 500 Lines
Data Party THU
Data Party THU
Jul 24, 2026 · Fundamentals

Maximum Likelihood Estimation Explained Simply: Core Principles, Code Walkthroughs, and Limitations

This article demystifies maximum likelihood estimation by using a coin‑flip example to contrast probability and likelihood, introduces log‑likelihood for numerical stability, shows Python implementations for discrete and continuous cases, discusses historical origins, and highlights practical limitations such as small‑sample bias and over‑fitting.

Log-LikelihoodMaximum Likelihood EstimationParameter Estimation
0 likes · 30 min read
Maximum Likelihood Estimation Explained Simply: Core Principles, Code Walkthroughs, and Limitations
Data Party THU
Data Party THU
Jul 23, 2026 · Artificial Intelligence

Why Adding Constraints Made This AI Coding Repo Go Viral

The article examines the popular GitHub repository mattpocock/skills, detailing how its four Claude Code skills—/grill-me, /tdd, /to-issues, and /diagnose—introduce deliberate friction into AI‑assisted programming, compares them with similar projects, and argues that the next wave of AI coding will prioritize disciplined, constraint‑driven workflows over raw speed.

AI codingClaudeGitHub
0 likes · 10 min read
Why Adding Constraints Made This AI Coding Repo Go Viral
Data Party THU
Data Party THU
Jul 22, 2026 · Artificial Intelligence

MemSlides Enables Precise Local Edits in AI‑Generated PPTs

MemSlides, a collaboration between Tsinghua, Shanghai Jiao Tong, and Beijing University of Posts and Telecommunications, introduces a hierarchical memory framework that lets a Slides Agent remember long‑term user preferences, temporary constraints, and tool experience, achieving accurate local PPT modifications and significantly improving personalization and edit efficiency.

AI PPTMemSlideshierarchical memory
0 likes · 13 min read
MemSlides Enables Precise Local Edits in AI‑Generated PPTs