Alyona Vert.

Alyona Vert.

Joined Turing Post in April 2024. Studied control systems of aircrafts at BMSTU (Moscow, Russia), where conducted several researchers on helicopter models. Now is more into AI and writing.

AI 101AI 101
Topic 46: RLHF variations: DPO, RRHF, RLAIF

Topic 46: RLHF variations: DPO, RRHF, RLAIF

we discuss three different methods for human alignment of LLMs that propose an alternative to the most widely-used RLHF

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 45: Reasoning Models - Just Advanced LLMs or New Species?

Topic 45: Reasoning Models - Just Advanced LLMs or New Species?

we discuss what defines a Reasoning Model, which ones are already in this category, and the potential paths to their future evolution

Ksenia Se
Alyona Vert.
Ksenia Se, +1
AI 101AI 101
Topic 44: The Human Touch: How HITL is Saving AI from Itself with Synthetic Data

Topic 44: The Human Touch: How HITL is Saving AI from Itself with Synthetic Data

we explore how human-in-the-loop systems are keeping synthetic data grounded, useful, and safe in the age of AI self-training

Alyona Vert.
Ksenia Se
Alyona Vert., +1
AI 101AI 101
Topic 43: What is Meta-Learning?

Topic 43: What is Meta-Learning?

look into the mind-bending world of meta-learning, where AI learns HOW to learn.

Ksenia Se
Alyona Vert.
Ksenia Se, +1
AI 101AI 101
Topic 42: Decoding BERT: From Original NLP Game-Changer to Today's Efficient AI (feat. ConstBERT)

Topic 42: Decoding BERT: From Original NLP Game-Changer to Today's Efficient AI (feat. ConstBERT)

we explore a foundational model architecture, BERT, which brought a deeper context understanding to reality and led to a wave of new context-aware models + the freshest ConstBERT, a multi-vector retrieval solution built on BERT

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 41: What's Cool About AlphaEvolve and Codex?

Topic 41: What's Cool About AlphaEvolve and Codex?

we discuss two fresh coding tools from Google DeepMind and OpenAI that reshape coding, making it more convenient and practical for real-world tasks optimization

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 40: What is GRPO and Flow-GRPO?

Topic 40: What is GRPO and Flow-GRPO?

we explore very clever algorithm behind DeepSeek-R1's breakthrough and how its latest adoption, Flow-GRPO make RL real for flow models

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 38: Can Liquid Models Beat Transformers? Meet Hyena Edge – the Newest Member of the LFM Family

Topic 38: Can Liquid Models Beat Transformers? Meet Hyena Edge – the Newest Member of the LFM Family

we discuss a new wave of architecture from Liquid AI – built from first principles, optimized for real hardware, and challenging the Transformer playbook with smarter, leaner models

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 37: What is MoE 2.0? Update Your Knowledge about Mixture-of-experts

Topic 37: What is MoE 2.0? Update Your Knowledge about Mixture-of-experts

The fresh angle on current Mixture-of-Expert. We discuss what new MoE techniques like S'MoRE, Symbolic-MoE, and others mean to the next generation AI

Alyona Vert.
Alyona Vert.
AI 101AI 101
Topic 36: What is Causal AI?

Topic 36: What is Causal AI?

we lift the curtain on what Causal AI is, its main principles, and how it can influence our world

Alyona Vert.
Alyona Vert.