41 Matching Annotations
  1. Last 7 days
    1. AlphaEvolve Design advanced algorithms for math and applications in computing

      This demonstrates the model's capacity for complex, structured problem-solving. To apply this, frame your prompts around a specific problem, provide all necessary constraints and requirements, and ask the model to design a step-by-step solution or algorithm.

  2. May 2026
    1. I also believe that the Scientist AI could even be more capable than the current approach, and that has to do with a number of design features. It is trained to explicitly reason in a structured way about the statements that it's asked to make a prediction over.

      Bengio大胆预测Scientist AI可能比现有方法更强大,因为它被训练以结构化方式推理,这一反直觉观点挑战了安全与能力必须取舍的假设,为安全AI提供了新视角。

  3. Apr 2026
    1. a stream of text that’s hard to hold onto, hard to compare, and hard to connect

      聊天界面的致命弱点在于缺乏结构,将所有输出压平为文本流,导致难以对比和关联。这解释了为何ChatGPT式交互适合探索却不适合严肃的团队协作——它把获得好结果的全部重担都压在了用户的提示词上。

  4. Dec 2025
    1. high segmentation significantly impacts cognitive load, vocabulary learning, retention,and reading comprehension across various aspects of multimedia learning. In essence, segmentation reducescognitive load, supports learning efficiency, and facilitates more profound understanding, vocabulary learning, andretention.

      Summary: This offers empirical proof that structure (segmentation) directly correlates with "learning efficiency" and "profound understanding," serving as the scientific backing for the "Professional Imperative" of standardization.

    2. segmenting dynamic visualizations intomeaningful units may aid learning by assisting learn-ers in grouping related elements and identifying naturalboundaries between events

      Summary: This explains how structure helps: it allows readers to identify "natural boundaries." This validates the use of standard grammar conventions as necessary markers that help the brain group and process ideas, especially for those still learning the English language.

    3. when essentialinformation is presented too rapidly, it can overload thelearner’s cognitive capacity, leading to cognitive overload.When this happens, the learner cannot process essentialinformation and learning outcomes effectively.

      Summary: Provides the consequence of poor structure: "cognitive overload." This supports the argument that unstructured or non-standard writing risks overloading the reader, preventing them from understanding the core message.

      Indirectly, this refutes the idea that "code-meshing" is necessary for more accurate communication.

  5. Sep 2025
  6. Mar 2025
    1. And for task-based concurrency you can adopt the smol model of task spawning to benefit from most of the benefits of structured concurrency today. And eventually the hope is we can add some form of async Drop to the language to close out the remaining holes.

      Yeah, this is a bit unfortunate. The TL;DR seems to be is that structured currency isn't fully possible in rust. smol seems to be the closest option, but it while it guarantees cancellation of child tasks eventually rust doesn't provided async destructors so there's no way to wait for the cancellations to complete before cleaning up the future.

  7. Nov 2024
  8. May 2024
  9. Sep 2022
    1. The authors propose, based on these experiences, that the cause ofa number of unexpected difficulties in human-computer interaction lies in users’ unwillingness orinability to make structure, content, or procedures explicit

      I'm curious if this is because of unwillingness or difficulty.

    1. https://www.kpcc.org/show/airtalk/2022-08-31/californias-legislative-session-ends-tonight-what-bills-are-on-the-chopping-block-as-time-winds-down

      Audio: https://traffic.omny.fm/d/clips/acc8cc57-ff7c-44c5-9bd6-ab0900fbdc43/39d4b39e-aa2d-40b1-bc07-ab2201235e43/fc58c5dd-b403-4747-a66c-af02013eab18/audio.mp3?utm_source=Podcast&in_playlist=f9e7dcf5-d162-4723-b710-ab2201235e4c

      Want to relisten to this. Caught the very end on interventions and it sounded very much like teaching "orality" rather than teaching literacy.

      Perhaps teaching orality first helps to frame literacy? even acknowledging it could help certainly...

      Cross reference LAist series: https://laist.com/news/education/dyslexia-california-teacher-preparation-training-structured-literacy

    2. California Could Mandate Kindergarten— What’s This Mean For School Districts And Childcare Providers?A bill that would create a mandatory kindergarten program in California has passed the legislature and is now heading to governor Gavin Newsom’s office for a final decision. The legislation, Senate Bill 70, would require children to complete one year of kindergarten before they’re admitted to the first grade. This comes as districts in California struggle with enrollment, having been a major issue during the pandemic. But if this legislation were to be signed by Governor Newsom, how would it affect teachers, the child care industry, and the children themselves.Today on AirTalk, we discuss the bill and it support among public schools with Los Angeles Unified School District (LAUSD) superintendent Alberto Carvalho and Justine Flores, licensed childcare provider in Los Angeles and a negotiation representative for Child Care Providers United.

      Timestamps 19:11 - 35:20

      CA Senate Bill 488 2021; signed, in process,

      Orton-Gillingham method (procedure/process) but can be implemented differently. Rigorous and works. Over 100 years old.

      Wilson program uses pieces of OG. What's this? Not enough detail here.

      Dyslexia training will be built into some parts of credentialling programs.

      Each child is different.

      This requires context knowledge on the part of the teacher and then a large tool bag of methods to help the widest variety of those differences.

      In the box programs don't work because children are not one size fits all.

      Magic wand ? What would you want?

      Madhuri would like to have: - rigorous teaching in early grades - if we can teach structured literacy following a specific scope in sequence most simple to most complex - teaching with same familiar patterns over and over - cumulative (builds on itself) - multisensory - explicit - Strong transitional kindergarten through grade 3 instruction

      Prevention trumps intervention.

      Otherwise you're feeding into the school to prison pipeline.

      Madhuri's call for teaching that is structured, cumulative, multisensory, and explicit sounds a lot like what I would imagine orality-based instruction looks like as well. The structure there particularly makes it easier to add pieces later on in a way that literacy doesn't necessarily.

  10. Jul 2022
    1. Kierkegaard has essentially this same view of human existence, a view that Becker praises in The Denial of Death. Because we are this tension of opposites, says Kierkegaard, in order to be authentically human we need to accept the mystery and responsibility of participation in both of these dimensions of reality that constitute life structured by death. Most people fall short of this authenticity, he declares. They flee its difficulties. And there are two basic ways of doing this. People either (1) immerse themselves in the dimension of things that perish, the things and pleasures of the world, which allows them to evade the awareness of death: the attitude summed up in the advice to “eat, drink, and be merry.” Or they (2) cling to some false certainty about immortality, imagining that some kind of immortality is their assured possession, and this too allows them to evade the awareness of death.

      Kierkegaard seems to look at death the same way as Becker. If we are authentic, it takes courage, first, but then we recognize it as wisdom. We participate in both the changing, perishable reality as well as the immutable, unchanging reality. Most people are too afraid to reach this point and evade a life structured by death in two major ways of denial of death. First they can live and let live. Enjoy all pleasures today with no regard for tomorrow. Second they can fall into an immortality project

  11. Jun 2022
  12. May 2022
  13. Nov 2021
  14. Jun 2021
  15. Mar 2021
  16. Feb 2021
  17. Jul 2020
    1. Valid AMP pages do not require schema.org structured data, but some platforms like Google Search require it for certain experiences like the Top stories carousel. It's generally a good idea to include structured data. Structured data helps search engines to better understand your web page, and to better display your content in Search Engine Result Pages (e.g., in rich snippets).
  18. May 2020
    1. Update 2020-01-14: I now store my outlines as Structure Zettel. For more information what a Structure Zettel is see this post.

      An important update to this piece as Sascha's method evolved. Instead of using outlines to capture new notes, he started using structured notes.

      I suspect the reason for this is that a system with atomic notes and structured notes is more clear cut than a system that relies on work-in-progress outlines. The main difference being that a structured note will contain only notes and not some floating, un-evolved ideas.

  19. Oct 2018
  20. Sep 2018
  21. Jun 2018
    1. Acero F., Ackermann M., Ajello M. et al (Fermi-LAT) 2015 arXiv:1501.02003Preprint

      Starting in 2014-2015, AAS/IOP started linking to preprints in reference lists if they were the version cited by the author and an accepted manuscript did not at that time exist.

      Thus we now have built in "categories" for references, which could be expanded to include data/software sections.

    1. 1- 13 A13 --- Planet Planet 15- 15 I1 --- robust Robust flag (1) 17- 23 F7.3 d Per Orbital period 25- 28 F4.1 Rgeo Rad Planet radius 30- 33 F4.2 Rgeo E_Rad 1{sigma} upper error bound on Rad 35- 38 F4.2 Rgeo e_Rad 1{sigma} lower error bound on Rad 40- 40 A1 --- r_Rad Source of planet-star radius ratio (2) 42- 44 F3.1 solMass Mstar Mass of host star 46- 49 F4.2 solMass E_Mstar 1{sigma} upper error bound on Mstar 51- 54 F4.2 solMass e_Mstar 1{sigma} lower error bound on Mstar 56- 56 I1 --- l_Md Md upper limit flag (3) 58- 63 F6.2 Mgeo Md Planet mass from default prior 65- 69 F5.2 Mgeo E_Md ?="" 1{sigma} upper error bound on Md 71- 74 F4.2 Mgeo e_Md ?="" 1{sigma} lower error bound on Md 76- 81 F6.2 g/cm3 rhod Planet density from default prior 83- 87 F5.2 g/cm3 E_rhod ?="" 1{sigma} upper error bound on rhod 89- 92 F4.2 g/cm3 e_rhod ?="" 1{sigma} lower error bound on rhod 94- 94 I1 --- l_Mh Mh upper limit flag (3) 96-100 F5.1 Mgeo Mh Planet mass from high mass prior 102-107 F6.2 Mgeo E_Mh ?="" 1{sigma} upper error bound on Md 109-112 F4.2 Mgeo e_Mh ?="" 1{sigma} lower error bound on Md 114-119 F6.2 g/cm3 rhoh Planet density from high mass prior 121-126 F6.2 g/cm3 E_rhoh ?="" 1{sigma} upper error bound on rhod 128-131 F4.2 g/cm3 e_rhoh ?="" 1{sigma} lower error bound on rhod 133-155 A23 --- Ref References (4)

      This is the main header block of the AAS Journal's "Machine Readable Format" for structured tables. It is based on the CDS table format, and follows their structuring rules. There are columns for the numerical format, units, labels, and explanations for each column.

  22. Jan 2016