TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Graph-based AI model maps the future of innovation

92 点作者 laurex6 个月前

15 条评论

abeppu6 个月前
Skimming the actual paper ... it seems pretty bad?<p>The thing about Beethoven&#x27;s 9th and biological materials which is mentioned in the OP is just that, out of a very large knowledge graph, they found small subgraph isomorphic to a subgraph created from a text about the symphony. But they seem not to cover the fact that a sufficiently large graph with some high-level statistical properties would have small subgraphs isomorphic to a &#x27;query&#x27; graph. Is this one good or meaningful in some way, or is it just an inevitable outcome of having produced such a large knowledge graph at the start? The reader can&#x27;t really tell, because figure 8 which presents the two graphs has such a poor resolution that one cannot read any of the labels. We&#x27;re just expected to see &quot;oh the nodes and their degrees match so it has the right shape&quot;, but that doesn&#x27;t really tell us that their system had any insight through this isomorphism-based mining process.<p>For the stuff about linking art (e.g. a Kandinsky painting) with material design ... they used an LLM to generate a description of a material for DALL-E where the prompt includes information about the painting, and then they show the resulting image and the painting. But there&#x27;s no measure of what a &quot;good&quot; material description is, and there certainly is no evaluation of the contribution of the graph-based &quot;reasoning&quot;. In particular an obvious comparison would be to &quot;Describe this painting.&quot; -&gt; &quot;Construct a prompt for DALL-E to portray a material whose structure has properties informed by this description of a painting ...&quot; -&gt; render.<p>It really seems like the author threw a bunch of stuff against the wall and didn&#x27;t even look particularly closely to see if it stuck.<p>Also, the only equation in the paper is the author giving the definition of cosine similarity, before 2 paragraphs justifying its use in constructing their graph. Like, who is the intended audience?<p><a href="https:&#x2F;&#x2F;iopscience.iop.org&#x2F;article&#x2F;10.1088&#x2F;2632-2153&#x2F;ad7228#mlstad7228t6" rel="nofollow">https:&#x2F;&#x2F;iopscience.iop.org&#x2F;article&#x2F;10.1088&#x2F;2632-2153&#x2F;ad7228#...</a>
评论 #42130025 未加载
评论 #42129409 未加载
评论 #42132430 未加载
评论 #42130519 未加载
评论 #42133423 未加载
gran_colombia6 个月前
&gt; One comparison revealed detailed structural parallels between biological materials and Beethoven’s 9th Symphony, highlighting shared patterns of complexity through isomorphic mapping.<p>This is not serious.
评论 #42129279 未加载
fudged716 个月前
Oh I&#x27;m glad that I&#x27;m not the only one who has gotten lost in the sauce by asking LLMs to recursively synthesize from data towards some grand insights--we want to see results when there is none apparent. What you end up getting is some bizarre theories overfit on the data with zero causal relationships. LLMs are fundamentally pattern matching systems and they will find &quot;connections&quot; between any two domains if prompted. It just reeks of confirmation bias; researchers looking for connections between art and science will find them.<p>The simpler explanation makes more sense: knowledge graphs naturally show certain structural properties, and these properties appear across domains due to basic mathematical constraints, common organizational principles, and human cognitive patterns reflected in data. Sure, LLMs trained on human knowledge can identify these patterns, generate plausible narratives, and create appealing connections - but this doesn&#x27;t necessarily indicate novel scientific insights, predictive power, or practical utility.<p>If you find yourself going down a rabbit hole like this (and trust me, we&#x27;ve all been there), my advice is to ask &quot;is there a simpler explanation that I&#x27;m missing?&quot; Then start from square one: specific testable hypotheses, rigorous controls, clear success metrics, practical demonstrations, and independent validation. And maybe add a &quot;complexity budget&quot; - if your explanation requires three layers of recursive AI analysis to make sense, you&#x27;re probably way too deep in the sauce.
youoy6 个月前
I think this article marks the &quot;peak of inflated expectations&quot; of AI for HN posts.
__reset__6 个月前
The author removed bridge construction from the <i>civil engineering</i> curriculum at MIT when he was heading the department (competitive steel bridge building is a big thing between CE departments in the US).<p>He said they were producing too many engineers and not enough scholars. When alumni offered to endow the program (in case it was a funding issue) he refused our donations.<p>Which makes this “scholarship” of chaining together some GPT prompts especially insulting.
quantadev6 个月前
Since all humans alive today have undergone the sum total of all human evolution, and are the ultimate creation of millions of years of evolution, it makes sense that the kinds of things we find &quot;artistically pleasing&quot; (both visually and thru sound) could have many patterns that apply to reality in deeper ways than any of us know, and so letting AI use art as it&#x27;s inspiration for using those patterns in it&#x27;s search for knew knowledge seems like a good idea.<p>Also there are also certain aspects of physical geometric relationships and even sound relationships that would not be able to be conveyed to an AI by any other means than thru art and music. So definitely using art to inspire science is a good approach.<p>Even the great Physicists throughout history have often appreciated how there is indeed beauty in the mathematical symmetries and relationships exhibited in the mathematics of nature, and so there is definitely a connection even if not quite tangible nor describable by man.
drawnwren6 个月前
Is it just me or does this read like complete word soup?<p>&gt; The application could lead to the development of innovative sustainable building materials, biodegradable alternatives to plastics, wearable technology, and even biomedical devices.<p>That a transform from materials to a 19th century Russian painter somehow is applicable to what just so happens to be the zeitgeist of materials science beggars belief.
nnurmanov6 个月前
Since the article mentions graphs, I’d like to ask what would be the advantages of graph databases over relational? Graph databases have become popular in RAG related topics, maybe mainly GraphRag related work by MS. So I wonder if the same accuracy with RAG could be achieved by traditional databases. Or if graph databases are an absolute must, then what are their limitations? Are there any successful production usage cases of graph databases?
woozyolliew6 个月前
One to save for April 1st
IAmGraydon6 个月前
We should probably flag this article out of existence as it&#x27;s pure garbage. Quite strange its getting enough upvotes to stay on the front page, but literally zero positive comments. The OP has an interesting history of posting lots of low quality articles.
评论 #42132831 未加载
331c8c716 个月前
Can we call this &quot;Deep Trolling&quot;?
评论 #42130446 未加载
quataran6 个月前
Wow, what&#x27;s happened to MIT?
评论 #42129960 未加载
byyoung36 个月前
did it actually make a novel material that is plausibly useful?
CatWChainsaw6 个月前
&quot;The Future of Innovation&quot; sounds exactly like freshly squeezed GPT drivel I&#x27;d expect to read from a vapid &quot;hustler&quot; on LinkedIn.
dacox6 个月前
...k