Daniel Weld (@dsweld) 's Twitter Profile
Daniel Weld

@dsweld

Computer science prof & entrepreneur. Excited by crowdsourcing, computational/social systems, MOOCs, relation extraction and Web-scale NLP.

ID: 24753328

linkhttp://www.cs.washington.edu/homes/weld/ calendar_today16-03-2009 19:39:33

1,1K Tweet

2,2K Followers

268 Following

Akari Asai (@akariasai) 's Twitter Profile Photo

3/ 🔍 What is OpenScholar? It's a retrieval-augmented LM with 1️⃣ a datastore of 45M+ open-access papers 2️⃣ a specialized retriever and reranker to search the datastore 3️⃣ an 8B Llama fine-tuned LM trained on high-quality synthetic data 4️⃣ a self-feedback generation pipeline

3/ 🔍 What is OpenScholar?
It's a retrieval-augmented LM with
1️⃣ a datastore of 45M+ open-access papers
2️⃣ a specialized retriever and reranker to search the datastore
3️⃣ an 8B Llama fine-tuned LM trained on high-quality synthetic data
4️⃣ a self-feedback generation pipeline
Daniel Weld (@dsweld) 's Twitter Profile Photo

Nice work on a method and UI for human-ai teaming! arxiv.org/abs/2412.10999 Task is the process of performing scientific research, but technique is general. With Kevin Feng Joseph Chee Chang Amy Zhang and many more

Ai2 (@allen_ai) 's Twitter Profile Photo

Introducing olmOCR, our open-source tool to extract clean plain text from PDFs! Built for scale, olmOCR handles many document types with high throughput. Run it on your own GPU for free—at over 3000 token/s, equivalent to $190 per million pages, or 1/32 the cost of GPT-4o!

Aaron Tay (@aarontay) 's Twitter Profile Photo

The trouble with finding a specific papers you read but forgetten is there's a diff between 1)what the paper says 2) what other people cite it for 3) what YOU want to cite it for! Trying ai2 paper finder makes it clear ha