Official platform of Tsinghua Big Data Research Center, sharing the team's latest research, teaching updates, and big data news.
This article experimentally evaluates a privacy‑preserving LLM security pipeline that discards raw dialogue after extracting 28 telemetry features, showing that using only 11 text‑independent signals retains about 98.5% of detection performance while reducing false‑positive rates.
The article analyzes the HyperAgents framework (DGM‑H), showing how merging task and meta agents enables metacognitive self‑modification, improves performance across coding and non‑coding benchmarks, automatically builds supporting infrastructure, and raises new safety and industry‑impact considerations.
MemPO introduces a self‑memory policy optimization framework that lets long‑horizon LLM agents autonomously manage and refine their memory via reinforcement learning, using global‑trajectory and informative‑memory advantage estimates, achieving up to 25.98% F1 gain and 73% token reduction on benchmark tasks.
The article reviews recent growth in AI agent marketplaces and systems, introduces a three‑dimensional framework—topology, memory scope, and update behavior—to categorize large‑scale multi‑agent networks, and highlights world‑model inconsistency as the core scalability bottleneck.
This comprehensive guide presents 21 practical text‑chunking techniques—from simple line‑based splits to advanced embedding‑ and LLM‑driven methods—explaining their implementations, code examples, and ideal use‑cases to help you build efficient Retrieval‑Augmented Generation systems while avoiding common pitfalls.
The MME-Emotion benchmark, introduced by researchers from CUHK and Alibaba Tongyi and accepted at ICLR 2026, provides a large‑scale, multimodal evaluation of emotional intelligence in large language models, revealing current models’ limited emotion recognition and reasoning abilities across diverse real‑world scenarios.
This article evaluates the top eleven backend programming languages for 2026, outlining each language's strengths, drawbacks, typical adopters, and offering a four‑question framework to help teams select the most suitable technology for their projects.
The article examines the rapid rise of OpenClaw, the open‑source AI agent dubbed “raising lobsters,” outlining its deployment steps, five major benefits, three key risks, and the six user profiles best suited for this execution‑type AI, while urging cautious adoption.
This paper introduces a novel framework that constructs and aligns dual visual‑textual trees on heterogeneous hyperbolic manifolds, addressing asymmetric modality alignment in hierarchical classification tasks and achieving state‑of‑the‑art performance on benchmarks such as CIFAR‑100, ImageNet and Rare Species datasets.
Discover ten essential Pydantic V2 techniques—including strict mode, field constraints, separate create/update/response models, cross‑field validators, custom error handling, reusable types, forbidden extra fields, nested models, computed fields, and discriminated unions—to prevent subtle bugs and ensure robust, secure FastAPI APIs in production.
