Primary Source
8 resources · page 1 of 1
Text-to-text prompt guide (Alibaba Cloud)
Learn to specify context, goals and output format using the model publisher’s own prompting guidance.
Indirect prompt injection in the real world: how people manipulate neural networks
by Vladislav Tushkanov (Kaspersky)
Read a Russia-origin security team’s field study of indirect prompt injection and the implications for defensive risk assessment.
Yandex Research: natural language processing
Follow primary Yandex research on language models, inference, distributed training and related datasets.
JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models
by Patrick Chao, Edoardo Debenedetti, Alexander Robey, Maksym Andriushchenko and colleagues
Learn why jailbreak studies need consistent threat models, scoring and reproducible comparisons.
GigaChat Family: Efficient Russian Language Modeling Through Mixture of Experts Architecture
by GigaChat team / SaluteDevices
Study a Russian-origin model team’s account of mixture-of-experts architecture, pretraining and evaluation.
HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
by Mantas Mazeika, Long Phan and colleagues
Study a standardized approach to red-team evaluation and robust refusal without copying attack recipes.
Qwen3 Technical Report
Read the model team’s account of Qwen3’s dense and MoE releases and unified thinking and non-thinking modes.
Qwen3Guard Technical Report
by Haiquan Zhao and colleagues / Qwen Team
Understand Qwen’s multilingual safety classifiers and the distinction between whole-response and streaming moderation.