Log inSign up
user avatar

Sam Buchanan

@_sdbuchanan
prev. @Berkeley_EECS, @TTIC_Connect, PhD @EE_ColumbiaSEAS. Training efficiency, theory and practice!
Bay Area, CA
sdbuchanan.com
Joined March 2015
1,408
Following
1,861
Followers
RepliesMedia
  • Pinned
    user avatar
    Sam Buchanan
    @_sdbuchanan
    Oct 2, 2025
    We wrote a book about representation learning! It’s fully open source, available and readable online, and covers everything from theoretical foundations to practical algorithms. 👷‍♂️ We’re hard at work updating the content for v2.0, and would love your feedback and contributions
    Cover of v1.0 of the book "Learning Deep Representations of Data Distributions", by Sam Buchanan, Druv Pai, Peng Wang, Yi Ma
  • user avatar
    Sam Buchanan
    @_sdbuchanan
    Mar 9
    We've released an updated "v2.0" of our book on deep representation learning! We've reorganized and improved many sections for better pedagogical clarity, and added many new examples and applications throughout the book. Massive thanks are due to folks in the community who
  • user avatar
    Sam Buchanan
    @_sdbuchanan
    Jan 12
    Escape the tyranny of the KV cache at large context lengths via end-to-end test-time training! I had the privilege to work with this team at the beginning of last year. The rigor and vision that went into this is remarkable (metalearning a transformer!?) -- check it out!
    user avatar
    Karan Dalal
    @karansdalal
    Jan 12
    LLM memory is considered one of the hardest problems in AI. All we have today are endless hacks and workarounds. But the root solution has always been right in front of us. Next-token prediction is already an effective compressor. We don’t need a radical new architecture. The
    Image
  • user avatar
    Sam Buchanan
    @_sdbuchanan
    Dec 4, 2025
    Presenting this morning at 11AM! We have a "laboratory" setting in which to study memorization+generalization in generative models. It allows the researcher to isolate different underlying factors, and characterize them theoretically and empirically. Come by and learn more!
    user avatar
    Druv Pai
    @druv_pai
    Dec 2, 2025
    Will be presenting this paper at NeurIPS 2025! 📅 Thursday, December 4, 11AM-2PM 📍 Exhibit Hall C, D, E #3703 DM me or come by in person if you want to chat about this work, or in general about representation learning, reasoning, generalization, and science of deep learning!
    Image