Hacker News

New stories

Live mirror
30 storiesupdated just nowView source snapshot
  1. Show HN: TopoTrace – Modernized, self-hosted IT asset and fleet risk tracker (topotrace.org)
    —discuss
  2. Brazilian President Lula clip dataset (huggingface.co)
    —discuss
  3. AI is deciding whether or not people receive medical care (vox.com)
    2comments
  4. Rails and AI (rubyonrails.org)
    —discuss
  5. Remembering Johannes Doerfert (llvm.org)
    —discuss
  6. SWE-Serve: Benchmarking Agentic Engineering for Production Inference Serving (nvidia.com)
    —discuss
  7. Propylea – Sub-Microsecond L7 Reverse Proxy and Active Defense in Rust (github.com/xuoxod)
    —discuss
  8. Stats: 72% of respondents said they felt curious, happy or excited about AI (marginalrevolution.com)
    —discuss
  9. Agentic ERP (shakegraph.com)
    —discuss
  10. Anthropic to Pay Akamai Technologies $11.6B over Seven Years for Cloud Services (wsj.com)
    —discuss
  11. Accomplishing More with LFM2.5-VL-DSpark (2tinteractive.com)
    1comments
  12. Mercedes-Benz gets first dibs on ProLogium's Gen4 solid-state EV battery cells (electrek.co)
    —discuss
  13. Valen: A multimodal decision model inspired by Jev (github.com/liuziyu77)
    —discuss
  14. The Murky History of Soviet-Born Tetris (mitpress.mit.edu)
    —discuss
  15. GitLab Outage (status.gitlab.com)
    10comments
  16. Show HN: Jev-browse – browser sub-tasks for coding agents at ~1/3 the cost (github.com/danielnc)
    —discuss
  17. It's sinister that Meta's Muse AI mascot is so cute (theverge.com)
    —discuss
  18. MentalHealthBench (openai.com)
    —discuss
  19. CTNicholas Multiplayer-Coding-Agents (github.com/ctnicholas)
    —discuss
  20. Special police to investigate threats to democracy, Swedish election fraud (kvartal.se)
    1comments
  21. Launching FreeBSD/EC2 Desktop AMIs (daemonology.net)
    —discuss
  22. Turbocharging Your Python Workflow (medium.com/therealcomtom)
    —discuss
  23. Award-Winning French Author Accused of Using AI to Write Debut Novel (cbc.ca)
    —discuss
  24. Zuse: Open-Source Cloud Agents (github.com/swarajbachu)
    —discuss
  25. Cal Newport: Here's What to Ask A.I. Leaders (nytimes.com)
    —discuss
  26. One API for Frontier Media Models (comfy.org)
    —discuss
  27. Welcome to the Age of Listening Devices (yahoo.com)
    1comments
  28. Interesting QA Career building book with interview tips (springer.com)
    —discuss
  29. Building an Ultra-High Throughput AI-SQL Engine (fsdatalab.github.io)
    —discuss
  30. Goldman reaped more than $200M in fees from hedge fund Situational Awareness (ft.com)
    —discuss

Any nerds out there who've read a lot of research papers?

4 pointsby 4h ago
9 comments
Which are the most memorable abstracts you have read? I know I am a jackass who usually vibecodes my way through assignments but even I don't want to write a cookie-cutter intro for my paper. I wanna write something that stays with the reader. What paper has stuck with you, what made it memorable?
2h agoHN ↗

I'm well into the thousands of research papers read. Understood a fair fraction of them. Couldn't tell you what the abstract was for any. The memorable parts are the ideas and the results within the paper.

Some titles stick. "Attention is all you need". "SSA is functional programming". Plus more people see the title than read the abstract. Only one I got to name myself was "Shared Memory Remote Procedure Calls" which I now realise is in exactly the same theme of given the title, you don't need the abstract or the contents.

1h agoHN ↗

Gotta have a look at yours! And I remember the name "Attention is all you need" too!

1h agoHN ↗

Hard to separate the memorability of the abstract from the underlying material.

IMO the most memorable papers contain some unexpected simplification: a complicated problem turns out to reduce to something much simpler [0], perhaps for a counterintuitive reason [1].

So a memorable abstract should advertise that the paper contains some cute little trick. This may not help you actually publish though

[0]: Attack the RLHF problem with a simple classification loss: https://arxiv.org/abs/2305.18290

[1]: Frustratingly Easy Meta-Embeddings: https://arxiv.org/abs/1804.05262