Chris Samarinas (@CSamarinas) header
Chris Samarinas (@CSamarinas) profile picture

CS PhD at CIIR @manningcics, founder of @NagetInc. Researcher in NLP & Information Retrieval. Search, nuggets, search.

United States
Joined March 2020
Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)Photo by Chris Samarinas (@CSamarinas)
We're introducing Naget 🔍 a search engine for agents, and the groundwork for millions more, built by everyone and connected to each other. 🧵👇
Search is changing 🔍 AI agents have fundamentally different information needs, yet today's technology wasn't built to meet them. Introducing Naget: the trusted information layer for agents 🤖 🚀 Now in early access
1
5
1
10
5,814
🚀 If you're building agents, join the early access waitlist at naget.com We'd love to give you free early access if Naget is solving your problems.
1
58
🎓 Naget is built by the Center for Intelligent Information Retrieval (CIIR), a frontier information retrieval lab for over three decades.
53
This baby is crawling 2 billion pages per month and hitting 1,400 tokens/s. The room stays at 30+°C (86°F) with a signature Founder Mode scent of ionized ozone and 'Eau de Silicon'. #WebSearch #LLMs #GPUPoor #Naget
1
6
328
Chris Samarinas retweeted
🤖➡️📉 Post-training made LLMs better at chat and reasoning—but worse at distributional alignment, diversity, and sometimes even steering(!) We measure this with our new resource (Spectrum Suite) and introduce Spectrum Tuning (method) to bring them back into our models! 🌈 1/🧵
5
48
3
193
70,180
Chris Samarinas retweeted
Excited to release new repo: nanochat! (it's among the most unhinged I've written). Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single, dependency-minimal codebase. You boot up a cloud GPU box, run a single script and in as little as 4 hours later you can talk to your own LLM in a ChatGPT-like web UI. It weighs ~8,000 lines of imo quite clean code to: - Train the tokenizer using a new Rust implementation - Pretrain a Transformer LLM on FineWeb, evaluate CORE score across a number of metrics - Midtrain on user-assistant conversations from SmolTalk, multiple choice questions, tool use. - SFT, evaluate the chat model on world knowledge multiple choice (ARC-E/C, MMLU), math (GSM8K), code (HumanEval) - RL the model optionally on GSM8K with "GRPO" - Efficient inference the model in an Engine with KV cache, simple prefill/decode, tool use (Python interpreter in a lightweight sandbox), talk to it over CLI or ChatGPT-like WebUI. - Write a single markdown report card, summarizing and gamifying the whole thing. Even for as low as ~$100 in cost (~4 hours on an 8XH100 node), you can train a little ChatGPT clone that you can kind of talk to, and which can write stories/poems, answer simple questions. About ~12 hours surpasses GPT-2 CORE metric. As you further scale up towards ~$1000 (~41.6 hours of training), it quickly becomes a lot more coherent and can solve simple math/code problems and take multiple choice tests. E.g. a depth 30 model trained for 24 hours (this is about equal to FLOPs of GPT-3 Small 125M and 1/1000th of GPT-3) gets into 40s on MMLU and 70s on ARC-Easy, 20s on GSM8K, etc. My goal is to get the full "strong baseline" stack into one cohesive, minimal, readable, hackable, maximally forkable repo. nanochat will be the capstone project of LLM101n (which is still being developed). I think it also has potential to grow into a research harness, or a benchmark, similar to nanoGPT before it. It is by no means finished, tuned or optimized (actually I think there's likely quite a bit of low-hanging fruit), but I think it's at a place where the overall skeleton is ok enough that it can go up on GitHub where all the parts of it can be improved. Link to repo and a detailed walkthrough of the nanochat speedrun is in the reply.
669
3,305
731
23,895
5,824,290
This paper is one of the most interesting works in IR the last 5+ years.
Instructions/reasoning are now everywhere in retrieval - we want embeddings to do it all! 🚀 But... is it even possible? 🤔 Turns out, it's not possible for single-vector models 😱 theoretically and empirically! To make it obvious we OSS a simple eval SoTA models flop on! 🧵
3
1
17
3,399
I'm sick of glorified API wrappers and Chromium reskins. If you want an early glimpse into agentic browser use, check out and contribute to the open-source nanobrowser Chrome extension: github.com/nanobrowser/nanob…
comet invites demand gives me the early gmail launch vibes. what an incredible product it was and comet is still not in the same leagues but feels special to have the company build something people really want.
4
624
Chris Samarinas retweeted
I’ll present our full paper "Bridging the Gap: From Ad-hoc to Proactive Search in Conversations" tomorrow (16 July) at @SIGIRConf #SIGIR2025, in the Conversational IR and Intelligent Agents session, MANTEGNA Platea, Floor 1, 10:30–12:30. Paper: dl.acm.org/doi/10.1145/37263… – at Padua, Veneto
1
4
1
24
2,164
📢 New paper on scaling test-time compute for document re-ranking Do you want to know how to train compact 2-3B models that can reach the performance of 70B+ LLMs in reasoning-intensive ranking? 📄Check out the distillation + RL recipe in our paper: arxiv.org/abs/2504.03947
1
4
22
2,282
Do you want a way to evaluate both factuality and topical coverage in LLMs? 📢 Excited to share our new paper: "Beyond Factual Accuracy: Evaluating Coverage of Diverse Factual Information in Long-form Text Generation" Introducing ICAT - a new interpretable evaluation framework!
1
1
1
9
3,060
Why does this matter? Traditional metrics struggle with long-form content evaluation. An LLM might be factually correct but miss crucial topic aspects. ICAT solves this by: • Breaking down texts into claims • Verifying factual accuracy • Measuring claim-topic alignment
1
1
415
Chris Samarinas retweeted
Come to SIGIR Session M3.2: Conversational IR and Recommendation to hear from @CSamarinas about proactive conversational search! #SIGIR2024
1
20
1,251
Join me for my presentations at #SIGIR2024 M3.1 RAG session July 15 4pm 1. Towards a Search Engine for Machines: Unified Ranking for Multiple Retrieval-Augmented Large Language Models 2. Optimization Methods for Personalizing Large Language Models through Retrieval Augmentation
1
1
16
897
If you are attending #SIGIR2024, come to our (@snbruch, @cosimorulli1, @rventurini_) talk in M1.3 on Seismic (efficient approx. sparse retrieval)! @snbruch drafted a nice blog post to describe the algorithm, w/ plenty of context: bruch.io/blog/publications/2…
1
11
779