AakashKumarNain/annotated_research_papers

This repo contains annotated research papers that I found really good and useful

2,798

148 commits

updated Aug 29, 2026

See the code

README

Annotate Research Papers

Alt Text

Why annotated papers?

Do you love reading research papers? Or do you want to read more research papers but find them intimidating? Or are you looking for annotated research papers that are much easier to understand?

If you are in any of the categories listed above, then you have arrived at the right place. I spend a lot of time reading papers. It is a crucial part of my ML work. If you want to do research or you want to be a better ML engineer, then you should read papers. This habit of reading papers will help you to remain updated with the field.

Note: I am a pen-paper guy. Nothing beats the pen-paper reading experience, but in the ongoing scenarios (pandemic, lockdown, etc.), I am not able to print the papers. Taking this as an opportunity to share my thought process, I will be sharing the annotated research papers in this repo. The order of the papers won't strictly be according to the timeline on arXiv. Sometimes I put a paper on hold and read it after a while.

PS: I cannot annotate all the papers I read, but you can expect all the interesting papers to be uploaded here.

Table of Contents

FieldCategoryAnnotated Paper
Computer VisionAdaptive Risk MinimizationAbstract
Axial DeepLabCodeAbstract
ConvNextCodeAbstract
EfficientNetsV2CodeAbstract
SupervisedFlow-edge Guided Video CompletionCodeAbstract
Is Batch Norm Unique?Abstract
Knowledge Distillation: A good teacher is patient and consistentCodeAbstract
RandConvCodeAbstract
PolylossCodeAbstract
Scaling Down Deep LearningCodeAbstract
Segment AnythingAbstract
Supervised Contrastive LearningCodeAbstract
Vision TransformerCodeAbstract
Gaze-LLECodeAbstract
Are all negatives created equal in contrastive instance discrimination?Abstract
Towards Domain-Agnostic Contrastive LearningAbstract
Self-SupervisedEmerging Properties in Self-Supervised Vision TransformersCodeAbstract
Decoder Denoising PretrainingAbstract
Masked Autoencoders CodeAbstract
SwavCodeAbstract
What Should Not Be Contrastive in Contrastive LearningAbstract
Vision Transformers need RegistersAbstract
NEPACodeAbstract
Semi-SupervisedCoMatchCodeAbstract
Diffusion ModelsUnderstanding Diffusion ModelsAbstract
On the Importance of Noise Scheduling for Diffusion ModelsAbstract
Emergent Correspondence from Diffusion ModelsAbstract
GANsCycleGanCodeAbstract
Interpretability and ExplainabilityWhat is being transferred in transfer learning?CodeAbstract
Explaining in StyleCodeAbstract
NLPDo Language Embeddings Capture Scales?Abstract
mSLAMAbstract
CrammingAbstract
Shortened LlamaAbstract
CoPEAbstract
LLMs cannot plan but can help planningAbstract
Mixture of A Million ExpertsAbstract
Agent Workflow MemoryCodeAbstract
What Matters for Model Merging at Scale?Abstract
Matryoshka QuantizationAbstract
Scaling Laws Are Unreliable for Downstream TasksAbstract
On the Theoretical Limitations of Embedding-Based RetrievalAbstract
Lightning OPDAbstract
EAGLEAbstract
SpeechSpeechStewAbstract
mSLAMAbstract
WhisperXCodeAbstract
MLLMsVCoder: Versatile Vision Encoder for MLLMsCodeAbstract
Sigmoid Loss for Image-Text PretrainingAbstract
MobileCLIPAbstract
MM1Abstract
Ferretv2Abstract
VisualFactCheckerAbstract
JanusFLowCodeAbstract
OthersMulti-Task Self-Training for Learning General RepresentationsAbstract
Decoder Denoising Pretraining for Semantic SegmentationAbstract

Community Contributions

Note: The annotated papers in this section are contributed by the community. As I cannot verify the annotation for each paper, I will lay out the guidelines for annotations so that every annotated paper has similar annotated sections.

annotations
deep-learning
machine-learning
research
research-paper

Contributors

AakashKumarNain

146 commits

23subbhashit

1 commits

deep-diver

1 commits

AakashKumarNain/annotated_research_papers

This repo contains annotated research papers that I found really good and useful

2,798

148 commits

updated Aug 29, 2026

See the code

README

Annotate Research Papers

Alt Text

Why annotated papers?

Do you love reading research papers? Or do you want to read more research papers but find them intimidating? Or are you looking for annotated research papers that are much easier to understand?

If you are in any of the categories listed above, then you have arrived at the right place. I spend a lot of time reading papers. It is a crucial part of my ML work. If you want to do research or you want to be a better ML engineer, then you should read papers. This habit of reading papers will help you to remain updated with the field.

Note: I am a pen-paper guy. Nothing beats the pen-paper reading experience, but in the ongoing scenarios (pandemic, lockdown, etc.), I am not able to print the papers. Taking this as an opportunity to share my thought process, I will be sharing the annotated research papers in this repo. The order of the papers won't strictly be according to the timeline on arXiv. Sometimes I put a paper on hold and read it after a while.

PS: I cannot annotate all the papers I read, but you can expect all the interesting papers to be uploaded here.

Table of Contents

FieldCategoryAnnotated Paper
Computer VisionAdaptive Risk MinimizationAbstract
Axial DeepLabCodeAbstract
ConvNextCodeAbstract
EfficientNetsV2CodeAbstract
SupervisedFlow-edge Guided Video CompletionCodeAbstract
Is Batch Norm Unique?Abstract
Knowledge Distillation: A good teacher is patient and consistentCodeAbstract
RandConvCodeAbstract
PolylossCodeAbstract
Scaling Down Deep LearningCodeAbstract
Segment AnythingAbstract
Supervised Contrastive LearningCodeAbstract
Vision TransformerCodeAbstract
Gaze-LLECodeAbstract
Are all negatives created equal in contrastive instance discrimination?Abstract
Towards Domain-Agnostic Contrastive LearningAbstract
Self-SupervisedEmerging Properties in Self-Supervised Vision TransformersCodeAbstract
Decoder Denoising PretrainingAbstract
Masked Autoencoders CodeAbstract
SwavCodeAbstract
What Should Not Be Contrastive in Contrastive LearningAbstract
Vision Transformers need RegistersAbstract
NEPACodeAbstract
Semi-SupervisedCoMatchCodeAbstract
Diffusion ModelsUnderstanding Diffusion ModelsAbstract
On the Importance of Noise Scheduling for Diffusion ModelsAbstract
Emergent Correspondence from Diffusion ModelsAbstract
GANsCycleGanCodeAbstract
Interpretability and ExplainabilityWhat is being transferred in transfer learning?CodeAbstract
Explaining in StyleCodeAbstract
NLPDo Language Embeddings Capture Scales?Abstract
mSLAMAbstract
CrammingAbstract
Shortened LlamaAbstract
CoPEAbstract
LLMs cannot plan but can help planningAbstract
Mixture of A Million ExpertsAbstract
Agent Workflow MemoryCodeAbstract
What Matters for Model Merging at Scale?Abstract
Matryoshka QuantizationAbstract
Scaling Laws Are Unreliable for Downstream TasksAbstract
On the Theoretical Limitations of Embedding-Based RetrievalAbstract
Lightning OPDAbstract
EAGLEAbstract
SpeechSpeechStewAbstract
mSLAMAbstract
WhisperXCodeAbstract
MLLMsVCoder: Versatile Vision Encoder for MLLMsCodeAbstract
Sigmoid Loss for Image-Text PretrainingAbstract
MobileCLIPAbstract
MM1Abstract
Ferretv2Abstract
VisualFactCheckerAbstract
JanusFLowCodeAbstract
OthersMulti-Task Self-Training for Learning General RepresentationsAbstract
Decoder Denoising Pretraining for Semantic SegmentationAbstract

Community Contributions

Note: The annotated papers in this section are contributed by the community. As I cannot verify the annotation for each paper, I will lay out the guidelines for annotations so that every annotated paper has similar annotated sections.

annotations
deep-learning
machine-learning
research
research-paper

Contributors

AakashKumarNain

146 commits

23subbhashit

1 commits

deep-diver

1 commits