BestHub
Sign in
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 3, 2026 · Artificial Intelligence

Kuaishou’s Recommendation-as-Generation Shatters Content Limits, Boosts Ads

RaG (Recommendation-as-Generation) redefines short‑video recommendation by predicting user interests, converting them into discrete semantic IDs, and generating personalized ads via a multi‑agent pipeline, achieving industrial‑scale deployment for over 400 million daily users and delivering a 1.87 % lift in ad revenue.

D-SIDsGenerative Recommendation ModelIndustrial Deployment
0 likes · 14 min read
Kuaishou’s Recommendation-as-Generation Shatters Content Limits, Boosts Ads
Data Party THU
Data Party THU
Jul 2, 2026 · Artificial Intelligence

Multi-Task Bayesian In-Context Learning: Transformers Adapt to New Priors

The ICML 2026 paper reframes in‑context learning as approximate Bayesian inference, introduces explicit prior datasets as a context prefix for Transformers, and demonstrates through synthetic and real‑world experiments that this multi‑task approach closely matches Bayesian oracles while offering fast, controllable inference.

ICML 2026In-Context LearningPrior Adaptation
0 likes · 15 min read
Multi-Task Bayesian In-Context Learning: Transformers Adapt to New Priors
Data Party THU
Data Party THU
Jul 1, 2026 · Artificial Intelligence

How Leading AI Labs Build and Use Claude Skills Effectively

The article reveals Anthropic’s internal approach to Claude Skills, detailing a nine‑category taxonomy, key principles such as focus and verification, practical writing guidelines, and strategies for scaling, governance, and composition, offering actionable insights for teams deploying Claude Code.

AIAnthropicAutomation
0 likes · 16 min read
How Leading AI Labs Build and Use Claude Skills Effectively
Data Party THU
Data Party THU
Jul 1, 2026 · Artificial Intelligence

How PageIndex Redefines RAG: Unpacking Its Structural Advantage Over Traditional Vector Retrieval

PageIndex introduces a non‑vector, reasoning‑based RAG approach that builds a hierarchical index from a document’s structure, lets large language models navigate to relevant sections, and delivers precise, citation‑rich answers, making it especially effective for long, well‑structured texts such as financial reports, legal contracts, and academic papers.

LLMPageIndexRAG
0 likes · 8 min read
How PageIndex Redefines RAG: Unpacking Its Structural Advantage Over Traditional Vector Retrieval
Data Party THU
Data Party THU
Jun 30, 2026 · Artificial Intelligence

Large-Scale Sign Language Datasets: Resources, Benchmarks, and Annotation Standards

This ACL 2026 survey systematically reviews over 120 publicly available sign‑language datasets covering 35 languages, analyzes their modalities, annotation inconsistencies, and benchmark limitations, and proposes a 24‑field datasheet to promote reproducible and comparable AI research in sign language recognition, translation, and generation.

AI researchannotation standardsbenchmarks
0 likes · 15 min read
Large-Scale Sign Language Datasets: Resources, Benchmarks, and Annotation Standards
Data Party THU
Data Party THU
Jun 30, 2026 · Artificial Intelligence

Do Video Generation Models Really Reason? A 303‑Question Benchmark Exposes Their Reasoning Gaps

The article introduces the MME‑CoF‑Pro benchmark, which uses 303 carefully crafted video‑reasoning samples across 16 categories to evaluate seven leading video generation models, revealing that current models lack true reasoning ability, that prompting can both help and hurt coherence, and that the new Reasoning Score aligns well with human judgments.

Artificial IntelligenceEvaluationMME-CoF-Pro
0 likes · 11 min read
Do Video Generation Models Really Reason? A 303‑Question Benchmark Exposes Their Reasoning Gaps
Data Party THU
Data Party THU
Jun 29, 2026 · Artificial Intelligence

Mapping LLM Reasoning: Paradigms, Methods, and Failure Modes in a Periodic Table

This 103‑page survey of over 300 recent papers organizes large language model reasoning into a periodic‑table framework, explains where reasoning emerges, categorizes 36 method families across six dimensions, critiques accuracy‑only evaluation, and outlines key open challenges such as fidelity, robustness, calibration, generalization, efficiency, and safety.

AI safetyEvaluationLLM reasoning
0 likes · 13 min read
Mapping LLM Reasoning: Paradigms, Methods, and Failure Modes in a Periodic Table
Data Party THU
Data Party THU
Jun 29, 2026 · Artificial Intelligence

Ensuring Safety in Real-World Reinforcement Learning: Tsinghua’s Safe Exploration Equilibrium Mechanism

The article reviews a Tsinghua University paper published in IEEE TPAMI 2026 that introduces a Safe Exploration Equilibrium (SEE) framework for real‑world reinforcement learning, proving convergence to a safety equilibrium, detailing a two‑step algorithm, and validating it on three classic control tasks with zero constraint violations and rapid region expansion.

ControlEquilibriumReal-World RL
0 likes · 8 min read
Ensuring Safety in Real-World Reinforcement Learning: Tsinghua’s Safe Exploration Equilibrium Mechanism
Data Party THU
Data Party THU
Jun 27, 2026 · Artificial Intelligence

AI and Chemists Co-Develop TYR Inhibitors via Dual-Track Optimization

The study presents a dual-track strategy that combines deep reinforcement‑learning‑driven de novo molecular generation with expert‑guided medicinal chemistry to discover and optimize TYR inhibitors, demonstrating how AI expands chemical space while chemists ensure synthetic feasibility, leading to potent candidates such as AI10‑m15 with strong anti‑melanogenesis activity.

AI-driven drug discoveryTYR inhibitorchemical space exploration
0 likes · 8 min read
AI and Chemists Co-Develop TYR Inhibitors via Dual-Track Optimization