Nishith Jain's picture

In a Training Loop 🔄

Nishith Jain

KingNish

·

AI & ML interests

AI is fun actually.

Recent Activity

new activity 6 days ago

innova-ai/YuE-music-generator-demo:Update requirements.txt

new activity about 2 months ago

KingNish/Instant-Image:Fix: Update deprecated APIs and dependencies

new activity about 2 months ago

KingNish/Realtime-FLUX:Fix: Update deprecated APIs and dependencies

View all activity

Organizations

upvoted a paper 3 months ago

OmniLottie: Generating Vector Animations via Parameterized Lottie Tokens

Paper • 2603.02138 • Published Mar 2 • 151

upvoted a changelog 3 months ago

Hugging Face Changelog

Find All Your Blog Drafts in One Place

Feb 2

• 50

upvoted an article 3 months ago

Article

GEM Image: Building an AI That Actually Gets Educational Diagrams Right

AIPreplabs

•

Feb 21

• 5

upvoted 4 papers 4 months ago

DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing

Paper • 2602.12205 • Published Feb 12 • 83

OPUS: Towards Efficient and Principled Data Selection in Large Language Model Pre-training in Every Iteration

Paper • 2602.05400 • Published Feb 5 • 355

Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting

Paper • 2601.02151 • Published Jan 5 • 115

On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification

Paper • 2508.05629 • Published Aug 7, 2025 • 189

upvoted an article 4 months ago

Article

Unlocking Agentic RL Training for GPT-OSS: A Practical Retrospective

LinkedIn

•

Jan 27

• 76

upvoted a paper 4 months ago

SAMTok: Representing Any Mask with Two Words

Paper • 2601.16093 • Published Jan 22 • 44

upvoted 2 articles 5 months ago

Article

How We Built a Semantic Highlight Model To Save Token Cost for RAG

zilliz

•

Jan 15

• 67

Article

The Optimal Architecture for Small Language Models

codelion

•

Dec 26, 2025

• 121

upvoted a paper 5 months ago

Nested Learning: The Illusion of Deep Learning Architectures

Paper • 2512.24695 • Published Dec 31, 2025 • 46

upvoted 4 articles 6 months ago

Article

Decoding Strategies in Large Language Models

mlabonne

•

Oct 29, 2024

• 113

Article

New in llama.cpp: Model Management

ggml-org

•

Dec 11, 2025

• 137

Article

Muon vs MuonClip vs Muon+AdamW for Fine-Tuning

KingNish

•

Dec 9, 2025

• 14

Article

Tensor Parallelism (TP) in Transformers: 5 Minutes to Understand

qgallouedec

•

Dec 4, 2025

• 69

upvoted 2 papers 6 months ago

PretrainZero: Reinforcement Active Pretraining

Paper • 2512.03442 • Published Dec 3, 2025 • 50

Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate

Paper • 2501.17703 • Published Jan 29, 2025 • 59

upvoted 2 collections 7 months ago

Kandinsky 5.0 Video Lite

Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality. • 9 items • Updated Dec 14, 2025 • 15

Kandinsky 5.0 Video Lite Diffusers

Kandinsky 5.0 Video Lite is a lightweight 2B model that generates up to 10-second SD videos from English and Russian prompts with high visual quality. • 8 items • Updated Feb 9 • 5