Hugging Face's logo
amitk17's picture
2 14 40

amitk17 PRO

sweatSmile
angelo98765432's profile picture muskallen's profile picture emerson8765431's profile picture
·
https://www.amitkatyayana.com/

AI & ML interests

Training LLMs • Prompt Engineering • Fine-tuning • Retrieval-Augmented Generation (RAG) • LoRA • NLP • Educational AI • Financial NLP • Chatbot Design • Model Deployment • Transformers • Open-Source Contributions

Recent Activity

reacted to Kseniase's post with 👍 about 5 hours ago
10 awesome advanced LoRA approaches Low-Rank Adaptation (LoRA) is the go-to method for efficient model fine-tuning that adds small low-rank matrices instead of retraining full models. The field isn’t standing still – new LoRA variants push the limits of efficiency, generalization, and personalization. So we’re sharing 10 of the latest LoRA approaches you should know about: 1. Mixture-of-LoRA-experts → https://huggingface.co/papers/2509.13878 Adds multiple low-rank adapters (LoRA) into a model’s layers, and a routing mechanism activates the most suitable ones for each input. This lets the model adapt better to new unseen conditions 2. Amortized Bayesian Meta-Learning for LoRA (ABMLL) → https://huggingface.co/papers/2508.14285 Balances global and task-specific parameters within a Bayesian framework to improve uncertainty calibration and generalization to new tasks without high memory or compute costs 3. AutoLoRA → https://huggingface.co/papers/2508.02107 Automatically retrieves and dynamically aggregates public LoRAs for stronger T2I generation 4. aLoRA (Activated LoRA) → https://huggingface.co/papers/2504.12397 Only applies LoRA after invocation, letting the model reuse the base model’s KV cache instead of recomputing the full turn’s KV cache. Efficient in multi-turn conversations 5. LiLoRA (LoRA in LoRA) → https://huggingface.co/papers/2508.06202 Shares the LoRA matrix A across tasks and additionally low-rank-decomposes matrix B to cut parameters in continual vision-text MLLMs 6. Sensitivity-LoRA → https://huggingface.co/papers/2509.09119 Dynamically assigns ranks to weight matrices based on their sensitivity, measured using second-order derivatives Read further below ↓ Also, subscribe to the Turing Post: https://www.turingpost.com/subscribe
reacted to Ethank01's post with 👀 about 5 hours ago
Discussing iMini Nano Banana – Practical AI Templates for Creative Workflows 🎙️I’ve been experimenting with imini nano banana recently, and the new batch of 100+ AI templates feels like a real shift in how we handle creative workflows. 🎨One example: the photo-to-hand-drawn sketch template. I tested it with a travel photo (a blonde girl in front of the Eiffel Tower) → the output was a clean, consistent hand-drawn sketch in seconds. Normally, this would take manual editing in Photoshop/Illustrator, but the AI template cut the process down to a single click. 💡From a workflow perspective, this could streamline tasks for: Designers: quick iterations on style variations Marketers: generating campaign visuals at scale Content creators: producing unique artwork without advanced editing skills 🖌️If anyone here has tested imini nano banana in professional projects, I’d like to hear how it fits into your pipeline. Full details: https://imini.com/nano-banana
reacted to salma-remyx's post with 🚀 about 5 hours ago
Trustworthy AI evals has been an industry challenge for the last few years, so what's missing? Causal Reasoning. Model based eval frameworks can't tell you if your changes actually improved user outcomes - you need to take a systems level approach. At Remyx, we’re building the intelligence layer for AI experimentation. Check out this example on how we start laying the scaffolding to launch controlled experiments to turn your hypotheses into insights on what drives performance for your application. Check out the latest at Remyx in our docs: https://docs.remyx.ai Try your first experiment today! https://engine.remyx.ai
View all activity

Organizations

MLX Community's profile picture Bitsandbytes Community's profile picture hectorlabs's profile picture