Undermind – Deep scientific paper search with adaptive LLMs
We can now build drastically higher quality search because we can use LLMs in algorithms that mimic a human's systematic research process, instead of just roughly recommending results based on semantic embeddings or term frequency. We built a deep search LLM pipeline that takes a few minutes to carefully search all the scientific literature. You describe your complex goal, as you would to a colleague. Then, we carefully search 200M+ papers. We classify the preliminary results with GPT-4. We then adapt the search goals based on relevant/irrelevant papers uncovered and continue searching,…
Undermind – Deep scientific paper search with adaptive LLMs launched on April 2, 2024 with 49 votes, #673 of 4,347 launches that month and more than 90% of that year's launches. The site was still online in September 2026, 2 years on.
In plain words
Undermind is a scientific paper search tool that uses large language models to systematically explore over 200 million academic papers. Users describe their research question conversationally, and the system conducts an adaptive multi-stage search, classifying results with GPT-4 and refining queries based on relevant and irrelevant papers found. It aims to provide more comprehensive results than traditional semantic search by mimicking how researchers manually conduct literature reviews. The tool can estimate search completion and is designed for scientists seeking thorough coverage of academic literature.
written from the facts on this page · September 2026
From the sources
Undermind finds the papers others miss, reading full texts and figures, citing every source. Build command of your field with a living library as you research.
In the maker’s words, at launch
We can now build drastically higher quality search because we can use LLMs in algorithms that mimic a human's systematic research process, instead of just roughly recommending results based on semantic embeddings or term frequency. We built a deep search LLM pipeline that takes a few minutes to carefully search all…
Does a similar job
all alternatives →
Building a web search engine from scratch with 3B neural embeddings2025 · blog.wilsonl.in · ▲699End-to-end deep dive of the project, spanning a large GPU cluster, distributed RocksDB, and terabytes of sharded HNSW.
Deep search of all ML papers2024 · app.undermind.ai · ▲109Built an automated system to run a deep search of ArXiv and carefully find all the precise papers that exist on a complex topic. It's different from simple RAG because it searches, classifies, and adapts based on relevant papers it uncovers, and then continues until it finds every paper on a topic (trying to mimic the human research process). Benchmarked 10x higher accuracy and total retrieval compared to Google Scholar for a median search (whitepaper on website). Also knows when it is complete, and misses virtually nothing (< 3% or so, once it's converged). Website has a free trial and a…
- IMI made a website to semantically search ArXiv papers2024 · papermatch.mitanshu.tech · ▲324
As a grad student (and an ADHDer), I had trouble doing literature review systematically. To combat this, I made a website that finds similar papers using the meaning of the thing I am looking for. I used MixedBread's [^1] embedding model to generate vectors from the abstracts. I store and search similar vectors using Milvus [^2] and finally use Gradio [^3] to serve the frontend. I update the vector database weekly by pulling the metadata dataset from Kaggle [^4]. To speed up the search process on my free oracle instance, I binarise the embeddings and use Hamming distance as a metric. I would…
- WSWeb search using a ChatGPT-like model that can cite its sources2022 · beta.sayhello.so · ▲318
We’ve trained a generative AI model to browse the web and answer questions/retrieve code snippets directly. Unlike ChatGPT, it has access to primary sources and is able to cite them when you hover over an answer (click on the text to go to the source being cited). We also show regular Bing results side-by-side with our AI answer. The model is an 11-billion parameter T5-derivative that has been fine-tuned on feedback given on hundreds of thousands of searches done (anonymously) on our platform. Giving the model web access lessens its burden to need to store a snapshot of human knowledge…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 20d ago · stillwet.art
PixVerse R2▲381A real-time world model you can explore and change
AI · 28d ago · world.pixverse.video
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, April 2024
the whole month →
Supabase▲2,325The Postgres developer platform is now generally available
Dev tools · 2024 · supabase.com
Cal.com Platform▲1,601Build your pixel-perfect booking experience with Atoms
Dev tools · 2024 · cal.com
deco.cx 2.0▲1,078Build web apps 10x faster with Deno, JSX, TS & Tailwind
Dev tools · 2024 · decocms.com
