Tags for Michał Chromiak's blog

3D Reconstruction 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

Adapter 1

Sat, 18 Mar 2023 LoRA: Fine-tuning a model by learning a small update

Adapters 1

Fri, 26 May 2023 QLoRA: How to fine-tune large language models with less memory

Agentic AI 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

application 1

Fri, 25 Aug 2017 The power of patterns – embrace the hurricanes.

Attention model 1

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

Autoencoder 1

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

basics 1

Fri, 01 Sep 2017 Neural Networks Primer

BERT 2

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

Brain Models 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

Brain Simulation 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

Calibration 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

CoLT5 1

Thu, 30 Mar 2023 CoLT5: Reading long documents with selective computation

Computational Neuroscience 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

Conditional Computation 1

Thu, 30 Mar 2023 CoLT5: Reading long documents with selective computation

Connectomics 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

CV 5

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Wed, 05 May 2021 🌀 MLP-Mixer: How image patches communicate without attention

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

Decision Models 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

Diffusion 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

Drosophila 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

Enterprise AI 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

ERNIE 1

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

FlyWire 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

GPT 2

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Graph Dynamics 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

Hebbian Plasticity 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

I-JEPA 1

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

Interpretability 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

Jev 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

k-NN 1

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

Language Model 1

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

Laya 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

Linear Attention 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

LLM 1

Sat, 18 Mar 2023 LoRA: Fine-tuning a model by learning a small update

LongT5 1

Thu, 30 Mar 2023 CoLT5: Reading long documents with selective computation

LoRA 1

Fri, 26 May 2023 QLoRA: How to fine-tune large language models with less memory

Low-Rank 2

Fri, 26 May 2023 QLoRA: How to fine-tune large language models with less memory

Sat, 18 Mar 2023 LoRA: Fine-tuning a model by learning a small update

Machine translation 1

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

MAE 1

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

Markov Decision Process 1

Tue, 01 Jun 2021 Decision Transformer: Unifying sequence modelling and model-free, offline RL

MDP 2

Tue, 01 Jun 2021 Decision Transformer: Unifying sequence modelling and model-free, offline RL

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

MetaAI 1

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

MLP 1

Wed, 05 May 2021 🌀 MLP-Mixer: How image patches communicate without attention

Multi-task Learning 1

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

Multimodal Models 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

N-gram 1

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

Neural Networks 1

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

NeuroMechFly 1

Sat, 19 Sep 2026 From a fly connectome to a digital fly: what the experiments show

NLP 3

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

NMT 1

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

Offline RL 1

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

Optimization 1

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

PatternRecognition 1

Fri, 25 Aug 2017 The power of patterns – embrace the hurricanes.

PEFT 1

Sat, 18 Mar 2023 LoRA: Fine-tuning a model by learning a small update

Perplexity 1

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

Pre-training 1

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

QLoRA 1

Fri, 26 May 2023 QLoRA: How to fine-tune large language models with less memory

Reinforcement Learning 2

Tue, 01 Jun 2021 Decision Transformer: Unifying sequence modelling and model-free, offline RL

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

Representation Learning 3

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

RL 2

Tue, 01 Jun 2021 Decision Transformer: Unifying sequence modelling and model-free, offline RL

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

Robotics 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

Scalability 2

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Self-Supervision (SSL) 3

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

seq2seq 1

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

Sequence transduction 1

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

Smoothing 1

Thu, 30 Nov 2017 Language models: From n-gram counts to neural next-token prediction

Spatial Intelligence 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

SSL 1

Sat, 17 Jun 2023 I-JEPA: Learning from images by predicting missing features

State-space Models 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

Structured Outputs 1

Thu, 17 Sep 2026 Typed Decision Models: Jev and Laya in Agentic AI

T5 1

Thu, 30 Mar 2023 CoLT5: Reading long documents with selective computation

Temporal-Difference Learning 1

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

Transformer 6

Thu, 30 Mar 2023 CoLT5: Reading long documents with selective computation

Tue, 01 Jun 2021 Decision Transformer: Unifying sequence modelling and model-free, offline RL

Wed, 05 May 2021 🌀 MLP-Mixer: How image patches communicate without attention

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding

Tue, 12 Sep 2017 The Transformer – Attention is all you need.

Transformers 1

Wed, 01 Oct 2025 Dragon Hatchling: A careful guide to BDH's graph-inspired state-space model

Value Functions 1

Sat, 01 May 2021 Reinforcement learning: A practical primer on agents, returns, values, and policies

Vision Transformer 1

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

ViT 4

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

Sun, 14 Nov 2021 🎭 Masked autoencoder (MAE) for visual representation learning. From the author of ResNet.

Wed, 05 May 2021 🌀 MLP-Mixer: How image patches communicate without attention

Mon, 03 May 2021 🦖 DINO: Self-supervised ViTs learn strong features and semantic structure

Weak-supervision 1

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

World Models 1

Sat, 05 Sep 2026 Atlas: How one model turns images into controllable 3D worlds

WSL 1

Tue, 18 Apr 2023 DINOv2: How self-supervised visual features transfer across tasks

XLNet 1

Tue, 30 Jul 2019 ERNIE 2.0: Continual multi-task pre-training for language understanding