limoce

9 ms·

1. 31.▲
   Fully Integrated 2D Flash Chip Unveiled(bioengineer.org)

   2 points by limoce 1y ago|0 comments
2. 32.▲
   Linux 6.18 UDP receive performance improved by 47%, under DDoS(git.kernel.org)

   33 points by limoce 1y ago|2 comments
3. 33.▲
   OpenAI 2025 ICPC Submissions(github.com)

   1 points by limoce 1y ago|0 comments
4. 34.▲
   Unveiling Silicon Art: Dieshots of Microchip Masterpieces(dieshot.com)

   3 points by limoce 1y ago|0 comments
5. 35.▲
   AetherCode: Evaluating LLMs' Ability to Win in Premier Programming Competitions(arxiv.org)

   1 points by limoce 1y ago|0 comments
6. 36.▲
   by limoce 1y ago

   "Train a 100,000,000,000B multimodal large language model" got 75/100
7. 37.▲
   Operation Costs in CPU Clock Cycles (2016)(ithare.com)

   32 points by limoce 1y ago|8 comments
8. 38.▲
   Principles and Methodologies for Serial Performance Optimization(usenix.org)

   1 points by limoce 1y ago|0 comments
9. 39.▲
   The Koala Benchmarks for the Shell(kben.sh)

   2 points by limoce 1y ago|0 comments
10. 40.▲
    SmallThinker: A Family of Efficient LLMs Natively Trained for Local Deployment(arxiv.org)

    2 points by limoce 1y ago|0 comments
11. 41.▲
    Step3 Technical Report [pdf](github.com)

    1 points by limoce 1y ago|0 comments
12. 42.▲
    FP8 is ~100 tflops faster when the kernel name has "cutlass" in it(twitter.com)

    243 points by limoce 1y ago|107 comments
13. 43.▲
    Polaris: A Post-training recipe for scaling RL on Advanced Reasoning models(hkunlp.github.io)

    4 points by limoce 1y ago|1 comments
14. 44.▲
    Overclocking LLM Reasoning: Monitoring and Controlling LLM Thinking Path Lengths(royeisen.github.io)

    63 points by limoce 1y ago|0 comments
15. 45.▲
    Neutrino: Probing-Based eBPF-Like GPU Kernel Profiling(github.com)

    2 points by limoce 1y ago|0 comments
16. 46.▲
    Machine Learning Conferences Should Establish "Refutations and Critiques" Track(arxiv.org)

    3 points by limoce 1y ago|0 comments
17. 47.▲
    by limoce 1y ago

    > https://preview.redd.it/wca7kzfq5w2f1.png?width=1190&format=... "4x gated residual streams" look quite weird. Is there any paper or technique report for this?
18. 48.▲
    SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines(supergpqa.github.io)

    2 points by limoce 2y ago|0 comments
19. 49.▲
    SepLLM: Accelerate LLMs by Compressing One Segment into One Separator(sepllm.github.io)

    39 points by limoce 2y ago|2 comments
20. 50.▲
    Step-Video-T2V: The Practice, Challenges, and Future of Video Foundation Model(arxiv.org)

    41 points by limoce 2y ago|5 comments
21. 51.▲
    Logic R1: Reproduce DeepSeek R1 Zero on 2K Logic Puzzle Dataset(github.com)

    1 points by limoce 2y ago|0 comments
22. 52.▲
    Libnginx: Nginx as a Shared Library(github.com)

    2 points by limoce 2y ago|0 comments
23. 53.▲
    DeepSeek-VL2: Moe Vision-Language Models for Advanced Multimodal Understanding [pdf](github.com)

    1 points by limoce 2y ago|0 comments
24. 54.▲
    Fast vectorizable algorithms of binary searching for floating point numbers(github.com)

    2 points by limoce 2y ago|0 comments
25. 55.▲
    New OpenAI Feature: Predicted Outputs(simonwillison.net)

    58 points by limoce 2y ago|7 comments
26. 56.▲
    Collaborative Filtering Is Wrong and Here Is Why(link.springer.com)

    2 points by limoce 2y ago|0 comments
27. 57.▲
    REST: A Plug-and-Play Method for Accelerating LLM Without Additional Training(sites.google.com)

    1 points by limoce 2y ago|0 comments
28. 58.▲
    Smoke 'em if you got 'em: Hacker gains root access using cigarette lighter(tomshardware.com)

    2 points by limoce 2y ago|0 comments
29. 59.▲
    O1 Replication Journey: A Strategic Progress Report(github.com)

    3 points by limoce 2y ago|0 comments
30. 60.▲
    Failures of Gradient-Based Deep Learning (2017) [pdf](proceedings.mlr.press)

    1 points by limoce 2y ago|0 comments