Ashish Chadha

Neural-Hacker

AshChadha-iitg

AI & ML interests

LLM , NLP , CV , TTS , GEN-AI

Recent Activity

upvoted a paper 38 minutes ago

Recursive Language Models

upvoted a paper 40 minutes ago

Prithvi-Complimentary Adaptive Fusion Encoder (CAFE): unlocking full-potential for flood inundation mapping

new activity about 14 hours ago

IndianAIDevs/README:Let's Talk about AI

View all activity

Organizations

upvoted a paper 38 minutes ago

Recursive Language Models

Paper • 2512.24601 • Published 8 days ago • 34

upvoted a paper 40 minutes ago

Prithvi-Complimentary Adaptive Fusion Encoder (CAFE): unlocking full-potential for flood inundation mapping

Paper • 2601.02315 • Published 3 days ago • 1

New activity in IndianAIDevs/README about 14 hours ago

Let's Talk about AI

#1 opened 5 months ago by

kalashshah19

upvoted a collection about 15 hours ago

LLM evals and benchmark datasets

Collection

16 items • Updated May 13, 2025 • 4

liked a dataset about 21 hours ago

JDhruv14/Bhagavad-Gita-QA

Viewer • Updated Oct 1, 2025 • 10.5k • 127 • 5

commented on Understanding Low-Rank Adaptation (LoRA): A Revolution in Fine-Tuning Large Language Models about 22 hours ago

Excellent question @AaTu9903 , and you've actually identified a genuine error in the article.
Since the LoRA adapter contribution is scaled by α/r, a larger alpha does amplify the adapter's effect, making the fine-tuning more pronounced, not less. It was a typing error from my side and I have fixed it. I really appreciate you catching that.