Log inSign up
Image
user avatar
Jimmy Lin
@lintool
I profess CS-ly at @UWaterloo. Previously, I monkeyed code for @Twitter, slides for @Cloudera, and scienced for @yupp_ai.
Nearby data lake
cs.uwaterloo.ca/~jimmylin/
Joined February 2010
864
Following
15.2K
Followers
  • user avatar
    Jimmy Lin
    @lintool
    Aug 8, 2020
    Reviewers automatically assume that simple is not novel. This is sheer laziness. Yes, it may be simple and obvious in retrospect, but someone had to have that insight first. Simple is good. Simple is robust, easy to implement and reproduce, broadly applicable, etc.
  • user avatar
    Jimmy Lin
    @lintool
    Jul 11, 2025
    It’s been 36 hours since Grok 4 launched and we have an early verdict based on 6K+ preferences of @yupp_ai users globally on real use cases. ‼️ Grok 4 is worse than other leading models: OpenAI o3, Claude Opus 4, and Gemini 2.5 Pro. Grok 4 is liked even less than Grok 3. 🧵
    Image
    328K
  • user avatar
    Jimmy Lin
    @lintool
    Oct 11, 2022
    DAAM... You saw it here first! Attribution maps for Stable Diffusion based on upscaling and aggregating cross-attention activations in the latent denoising subnetwork: arxiv.org/abs/2210.04885 example for "an angry, bald man doing research" below - demo at daam.ralphtang.com:8080
    Image
  • user avatar
    Jimmy Lin
    @lintool
    Mar 20, 2023
    GPT-4 and its ilk are awesome for rapid prototyping and one-offs, but at the end of the day, enterprises will deploy far smaller distilled models in production. Here's my contrarian take -
    207K
  • user avatar
    Jimmy Lin
    @lintool
    May 19, 2021
    So, CV researchers are looking at transformers and NLP researchers are looking at CNNs (again). What a strange world.
  • user avatar
    Jimmy Lin
    @lintool
    Jul 21, 2020
    Still cropping and modifying BERT diagrams from Devlin et al. (2019)? I spent several hours redrawing BERT in PowerPoint so you don't have to... Perfect for use in presentations, papers, etc.! Happy to share under Releasing under CC BY 4.0 cs.uwaterloo.ca/~jimmylin/BERT…
    Image
  • user avatar
    Jimmy Lin
    @lintool
    Feb 20, 2018
    Following the AI Residency program by Google, Facebook, Microsoft, Uber, etc., I'd like to start the Waterloo AI Residency program. It's called grad school.
  • user avatar
    Jimmy Lin
    @lintool
    Nov 24, 2021
    Look what came in the mail!
    Image
  • user avatar
    Jimmy Lin
    @lintool
    Jul 27, 2020
    "NLP makes IR interesting and IR makes NLP useful!" - slides from my #sigir2020 summer school talk at: cs.uwaterloo.ca/~jimmylin/publ… Get your rotten tomatoes and eggs out!
  • user avatar
    Jimmy Lin
    @lintool
    Oct 11, 2021
    BERT is three years old today!
    arXiv logo
    arxiv.org
    BERT: Pre-training of Deep Bidirectional Transformers for Language...
    We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, BERT is...
  • user avatar
    Jimmy Lin
    @lintool
    Oct 14, 2020
    Happy to share an early draft of "Pretrained Transformers for Text Ranking: BERT and Beyond", our forthcoming book (tentatively, early 2021) by @lintool @rodrigfnogueira @andrewyates
    arXiv logo
    arxiv.org
    Pretrained Transformers for Text Ranking: BERT and Beyond
    The goal of text ranking is to generate an ordered list of texts retrieved from a corpus in response to a query. Although the most common formulation of text ranking is search, instances of the...
  • user avatar
    Jimmy Lin
    @lintool
    Sep 11, 2024
    For vector search, practitioners kinda know that for small corpora, don't bother with HNSW indexing, just brute-force it. However, guidance is mostly hand wavy... until now. I ran some experiments for you on BEIR and wrote it up. arxiv.org/abs/2409.06464 You're welcome.
    25K
  • user avatar
    Jimmy Lin
    @lintool
    Feb 8, 2023
    My (contrarian?) take: prompt engineering is programming in natural language. We've tried this before, with attempts dating back decades. Recent advances do not change the fact that natural languages are ambiguous, imprecise, under-specified, highly contextual, etc.
    33K
  • user avatar
    Jimmy Lin
    @lintool
    Jan 16, 2023
    Recently, @CohereAI boasted "3X better performance" in multilingual text understanding. We tested that claim by evaluating Cohere embeddings on MIRACL: tl;dr - We weren't able to replicate the 3X claim, but we did observe a 38% improvement over BM25. github.com/castorini/pyse…
    62K