Deschide articolul original într-un tab nou Schimbarea acestei opțiuni se aplică imediat și reîncarcă pagina.
LECTURA
Înapoi la articole →

Juncheng Yang

blog.jasony.me/feed.xml
Vizitează site-ul

Cel mai recent articol

Ultima verificare

Urmăritori 0 urmăritori
Articole: 9

  1. A paper reproducibility agent: PDF in, report out

    (Juncheng Yang)

    We turned the MLSys'26 AE agent's core into a standalone paper reproducibility agent.

  2. Same open weights, prompt-cache lifetimes from seconds to a day

    (Juncheng Yang)

    Across 23 providers serving two open-weight models, prompt-cache retention ranged from under 30 seconds to more than a day.

  3. Deploying an artifact-evaluation agent at MLSys'26

    (Juncheng Yang)

    We deployed a tool-using agent on the MLSys'26 AE HotCRP as a real reviewer.

  4. Measure first, design second

    (Juncheng Yang)

    AI is making solutions cheaper, but it is not making important problems easier to find.

  5. Proprietary prompt caches expire in minutes, not hours — cheap per token doesn't mean a cheap bill

    (Juncheng Yang)

    OpenAI, Gemini, Anthropic, and xAI all cache your prompt prefix — but the probe that caught open-weight backends holding prefixes for hours catches the big proprietary APIs letting go in 5 to 30 minutes.

  6. Authorship inflation at OSDI: counting papers two ways

    (Juncheng Yang)

    OSDI author lists have more than doubled since 1994.

  7. Make model precision elastic

    (Juncheng Yang)

    MorphServe temporarily converts selected model layers into KV-cache headroom when traffic spikes, then restores full precision when the pressure passes.

  8. Why FIFO is (almost) all you need for cache eviction

    (Juncheng Yang)

    LRU tries to rank everything.

  9. What I look for in systems research students

    (Juncheng Yang)

    I am looking for curious, persistent students who want to build real systems.