vishwajeet kumar @vishwajeet_86

Research Scientist at IBM Research AI | Interested in Natural Language Processing and Machine Learning | IIT Bombay Joined June 2010

Tweets

2K
Followers

124
Following

2K
Likes

999

Rohan Paul @rohanpaul_ai

3 weeks ago

Github 👨‍🔧: Awesome-GraphRAG A curated list of resources (surveys, papers, benchmarks, and opensource projects) on graph-based retrieval-augmented generation. github .com/DEEP-PolyU/Awesome-GraphRAG

5 80 396 27K 432

Download Image

I implemented GRPO and DPO from scratch in vanilla Pytorch to unravel every piece of training details. Hope it could be helpful for those who care about the implementation details of the algorithms. 👉 github.com/mingyin0312/RL… #AI #RL #LLM

15 212 2K 102K 2K

ICLR 2026 @iclr_conf

a month ago

Announcing the ICLR 2026 Call for Papers! Abstract submission: Sept 19 (AoE) Paper submission: Sept 24 (AoE) Reviews released: Nov 11 Author/Reviewer discussion: Nov 11-Dec 3 Final decisions: Jan 22 2026 iclr.cc/Conferences/20…

3 63 534 47K 157

Victoria Slocum @victorialslocum

a month ago

ColBERT and ColPali are amazing. But they're also bankrupting your infrastructure budget. Remember when multi-vector embeddings meant choosing between quality and your AWS bill? 𝗠𝗨𝗩𝗘𝗥𝗔 (Multi-Vector Retrieval via Fixed Dimensional Encodings) 𝗵𝗮𝘀 𝗺𝗮𝗱𝗲 𝗶𝗻𝘀𝗮𝗻𝗲…

0 53 282 17K 245

Download Image

OpenAI @OpenAI

a month ago

Our open models are here. Both of them. openai.com/open-models

1K 3K 20K 6.5M 4K

elvis @omarsar0

a month ago

Claude Code Templates There is no time-wasting in this community. This Claude Code template collection looks really cool. You probably need to customize them further, but it's a good place to find inspiration.

22 92 817 85K 1K

Download Image

Ruben Hassid @RubenHssd

a month ago

NVIDIA just dropped paper exposing a $57 billion AI industry mistake. While Big Tech keeps pushing expensive LLMs like ChatGPT & Claude... Small language models handle 70% of AI agent work at 1/30th the cost. Here's why this changes everything: (hint: less is more)

58 252 1K 198K 2K

Download Image

Rohan Paul @rohanpaul_ai

a month ago

The paper argues the next big thing is the Agentic Web where autonomous AI agents, not humans, run tasks across the internet. The authors show how the Web has already shifted from early pages to mobile feeds and now toward agent‑driven action. They define 3 dimensions,…

20 131 543 40K 480

Download Image

Rohan Paul @rohanpaul_ai

a month ago

Wow. This is a HUGE 24-trillion-token web dataset with document-level metadata available on @huggingface apache-2.0 license - collected from Common Crawl - each document is labeled with a 12-field taxonomy covering topic, page type, complexity, and quality . - Labels are…

3 47 273 31K 178

Download Image

Simo Ryu @cloneofsimo

a month ago

@jerryx314 jerryxio.ng/posts/nd-rope/

0 4 37 3K 30

Rohan Paul @rohanpaul_ai

a month ago

Beautiful @GoogleResearch paper. LLMs can learn in context from examples in the prompt, can pick up new patterns while answering, yet their stored weights never change. That behavior looks impossible if learning always means gradient descent. The mechanisms through which this…

60 346 3K 360K 3K

Download Image

Avi Chawla @_avichawla

2 months ago

The only MCP server you'll ever need! MindsDB lets you query data from 200+ sources, like Slack, Gmail, social platforms, and more, in both SQL and natural language. A federated query engine that comes with a built-in MCP server. 100% open-source with 33k+ stars!

29 281 2K 206K 4K

Download Image

Rohan Paul @rohanpaul_ai

3 months ago

A 24-trillion-token web dataset with document-level metadata just dropped on @huggingface License: apache-2.0 ESSENTIAL-WEB v1.0 collects 24 trillion tokens from Common Crawl. Each document is labeled with a 12-field taxonomy covering topic, page type, complexity, and quality…

16 114 586 63K 501

Download Image

Tanishq Mathew Abraham, Ph.D. @iScienceLuvr

4 months ago

abs: arxiv.org/abs/2505.12082

0 3 30 5K 23

Kevin Patrick Murphy @sirbayes

4 months ago

I am pleased to announce a new version of my RL tutorial. Major update to the LLM chapter (eg DPO, GRPO, thinking), minor updates to the MARL and MBRL chapters and various sections (eg offline RL, DPG, etc). Enjoy! arxiv.org/abs/2412.05265

24 447 2K 119K 2K

Download Image

Haruki Sakajo @sjh4i

4 months ago

Our paper, Dictionaries to the Rescue: Cross-Lingual Vocabulary Transfer for Low-Resource Languages Using Bilingual Dictionaries, has been accepted to #ACL2025NLP Findings! Thanks to the co-authors, @Yusuke196 , Justin, @yusuke1997 , @alanyttian , @h_kamiga , @tarowatanabe !

1 5 36 4K 1

Shivalika Singh @singhshiviii

4 months ago

Super thrilled to share GMMLU is accepted to #ACL2025 main conference 🎉 It was also recently recognised by @StanfordHAI as one of the significant AI releases of 2024 🚀 I had a blast collaborating on this closely with @beyzaermis and all our collaborators! Huge congrats!💙

Shivalika Singh @singhshiviii

9 months ago

2 38 143 33K 48

Download Image

6 15 74 5K 5

NeurIPS Conference @NeurIPSConf

4 months ago

📢 A reminder that the NeurIPS deadline for full paper submission is May 15th (Anywhere on Earth, AOE). We look forward to receiving your work, good luck to all submitting authors!

13 11 236 95K 17

Taiwei Shi @taiwei_shi

4 months ago

Want to 𝐜𝐮𝐭 𝐑𝐅𝐓 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐭𝐢𝐦𝐞 𝐛𝐲 𝐮𝐩 𝐭𝐨 𝟐× and boost performance? 🚀 Meet 𝑨𝒅𝒂𝑹𝑭𝑻 — a lightweight, plug-and-play curriculum learning method you can drop into any mainstream RFT algorithms (PPO, GRPO, REINFORCE). Less compute. Better results. 🧵 1/n

3 55 359 43K 363

Download Image

elvis @omarsar0

4 months ago

Universal RAG RAG is dead, they said. Then you see papers like this and it gives you a better understanding of the opportunities and challenges ahead. Lots of great ideas in this paper. I've summarized a few below: