Badri N. Patro
@badripatro_iitb
Followers
265
Following
4K
Media
13
Statuses
2K
Senior Research Scientist @Microsoft | Prev: @Google, @Microsoft, @Samsung, @HARMAN | | PDF @KU_Leuven | PhD @IITKanpur | https://t.co/cNrlVQCXxw. @iitbombay
India
Joined October 2015
Home Maintenance Hacks: DIY Skills Repair Technicians Don't Want You to Know!👨🔧 Tag your hubby and let him in on these handy tips🧤👇👇
15
839
3K
Transformers vs Mixture of experts in LLMs visually explained:
Transformer vs. Mixture of Experts in LLMs, clearly explained (with visuals):
6
372
2K
#ICCV2025 reminders - All qualified authors are required to act as reviewers. - Any reviewer who fails to submit their assigned reviews by the deadline will face a desk rejection of all papers on which they are an author per the discretion of the PCs. https://t.co/rBRTjduBcx
1
5
47
दक्षिण भारत में बनी एक फिल्म का सीन है ! आपकी आँखें फटी की फटी रह जाएँगीं, बॉलीवुड तो ऐसी फिल्म क्या.. इस सीन जैसा भी कभी कुछ नहीं दिखा पाएगा ! क्योंकि वो बॉलीवुड नहीं रहा👍 पूरी दुनिया में दिखाने लायक सीन फिल्माया गया है। जो 100%सही है। सभी देखे और भेजे आगे से आगे।आपका आभार
204
4K
11K
CPU is all you need
Complete hardware + software setup for running Deepseek-R1 locally. The actual model, no distillations, and Q8 quantization for full quality. Total cost, $6,000. All download and part links below:
15
35
429
If you want to speak like a leader, watch this ‼️‼️
14
652
3K
इस शख्स को इतनी भयंकर जानकारी आप सुनकर दंग रह जाएंगे। एक बार जरूर सुने !
573
7K
19K
Glad to announce our paper named “SiMBA-TS: Looking Beyond Transformers for Long Term Time-Series Forecasting” which has been accepted in the ICASSP 2025 ( https://t.co/SWHvbABscw) A short summary of the paper: Thanks @badripatro_iitb, my co-author https://t.co/wgRecgyuEB
lnkd.in
This link will take you to a page that’s not on LinkedIn
0
1
1
📄🤖 PDF Q&A with Deepseek Create a production-ready AI chatbot that answers questions from PDFs using DeepSeek's LLM and LangChain's document processing. Features intelligent text processing, advanced reasoning, and an intuitive interface. https://t.co/pJY6YB7ZX1
18
114
605
Perplexity has announce free Perplexity Pro for the below colleges, if you belong to them, do get your free Pro - Indian Institute of Technology, Kharagpur - Indian Institute of Technology, Kanpur - Indian Institute of Technology, Delhi - Indian Institute of Technology,
251
274
4K
10x speed looks wild Deepseek R1 coder + OS OpenAI Operator Agent
DeepSeek-R1 coder + OS OpenAI Operator Agent (browser-use) prompt: Go to https://huggingfacedotco/spaces/akhaliq/anychat click deepseek coder in dropdown, type write a script for a bouncing yellow ball within a sphere, make sure to handle collision detection properly. then
11
71
514
DeepSeek built a high performance computer on Aug 31 with 10,000 A100 GPUs. A must-read paper only cited ONCE. In their V3 paper, the base for R1, they say they train on 2048 H800s, the export-controlled H100 with 50% the transfer rate. Why didn't they use the 10,000 A100s?
88
326
3K
10). Agentic RAG Overview Provides a comprehensive introduction to LLM agents and Agentic RAG. It provides an exploration of Agentic RAG architectures, applications, and implementation strategies. https://t.co/x1dCnZgbyc
Agentic RAG Overview This is a great intro to LLM agents and Agentic RAG. It provides a comprehensive exploration of Agentic RAG architectures, applications, and implementation strategies.
0
4
17
Here are the top AI Papers of the Week (Jan 20-26): - DeepSeek-R1 - Can LLMs Plan? - Chain-of-Agents - Scaling RL with LLMs - Humanity’s Last Exam - Agentic RAG Overview Read on for more:
10
176
1K
Transformer²: Self-adaptive LLMs https://t.co/VnvzSmeIad This new paper from @SakanaAILabs shows the power of an LLM that can self-adapt its weights to its environment. I think in the future, the line between “pre-training” and “post-training” will be gone, and our models and
We’re excited to introduce Transformer², a machine learning system that dynamically adjusts its weights for various tasks! https://t.co/ci028qPUWt Adaptation is a remarkable natural phenomenon, like how the octopus can blend in with its environment, or how the brain rewires
24
170
1K
[Download 698-page PDF eBook] Everything You Always Wanted To Know About #Mathematics* (*But didn’t even know to ask) A Guided Journey Into the World of Abstract Mathematics, Theorems, and the Writing of Proofs: https://t.co/JLsDOmpP1q
4
242
1K
[Download 496-page PDF eBook] Applied Causal #Inference Powered by #MachineLearning and #AI: https://t.co/NPkNAHaB47 ———— #ML #DataScience #Algorithms #Statistics
5
114
535
Pandas is a powerful data analysis and manipulation library for Python! NVIDIA just made Pandas 150x faster with zero code changes 🔥 All you have to add is just a couple of lines of code: %load_ext cudf.pandas import pandas as pd
15
197
1K
Damnnnn this tutorial is of high quality!! Check out the code repository: - Standalone Jupyter notebooks. - Only what you need. - Highly explanatory. - 0 over engineering. - All from scratch. - Self contained. I wish everyone wrote all their code like this.
Direct Preference Optimization (DPO) has become one of the go-to methods to align large language models (LLMs) more closely with user preferences. If you want to learn how it works, I coded it from scratch: https://t.co/VioT1zVn68
0
42
263