TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

KAG – Knowledge Graph RAG Framework

230 pointsby taikon5 months ago

14 comments

isoprophlex5 months ago
Fancy, I think, but again no word on the actual work of turning a few bazillion csv files and pdf&#x27;s into a knowledge graph.<p>I see a lot of these KG tools pop up, but they never solve the first problem I have, which is actually constructing the KG itself.
评论 #42547488 未加载
评论 #42549911 未加载
评论 #42549856 未加载
评论 #42552272 未加载
评论 #42547556 未加载
评论 #42550327 未加载
评论 #42551738 未加载
评论 #42562692 未加载
评论 #42548481 未加载
评论 #42547743 未加载
评论 #42549416 未加载
djoldman5 months ago
&quot;Whitepaper&quot; is guarded behind this: <a href="https:&#x2F;&#x2F;survey.alipay.com&#x2F;apps&#x2F;zhiliao&#x2F;n33nRj5OV" rel="nofollow">https:&#x2F;&#x2F;survey.alipay.com&#x2F;apps&#x2F;zhiliao&#x2F;n33nRj5OV</a><p>&gt; The white paper is only available for professional developers from different industries. We need to collect your name, contact information, email address, company name, industry type, position and your download purpose to verify your identity...<p>That&#x27;s new.
评论 #42551586 未加载
评论 #42549158 未加载
zbyforgotp5 months ago
LLMs are not that different from humans, in both cases you have some limited working memory and you need to fit the most relevant context into it. This means that if you have a new knowledge base for llms it should be useful for humans too. There should be a lot of cross pollination between these tools.<p>But we need a theory on the differences too. Now it is kind of random how we differentiate the tools. We need ergonomics for llms.
评论 #42548440 未加载
评论 #42548228 未加载
mentalgear5 months ago
It has come to the point that we need benchmarks for (Graph)-Rag systems now, same as we have for pure LLMs. However vendors will certainly then optimize for the popular ones, so we need a good mix of public, private and dynamic eval datasets.
swyx5 months ago
advice to OP - that gif showing how you zoom in and star the repo is a giant turnoff. i closed my tab when i saw that.
评论 #42549890 未加载
评论 #42552994 未加载
评论 #42547831 未加载
slowmovintarget5 months ago
How does this compare to the Model Context Protocol?<p><a href="https:&#x2F;&#x2F;modelcontextprotocol.io&#x2F;introduction" rel="nofollow">https:&#x2F;&#x2F;modelcontextprotocol.io&#x2F;introduction</a>
评论 #42547285 未加载
mentalgear5 months ago
I like their description&#x2F;approach for logical problem solving:<p>2.2.<p>&quot;The engine includes three types of operators: planning, reasoning, and retrieval, which transform natural language problems into problem solving processes that combine language and notation.<p>In this process, each step can use different operators, such as exact match retrieval, text retrieval, numerical calculation or semantic reasoning, so as to realize the integration of four different problem solving processes: Retrieval, Knowledge Graph reasoning, language reasoning and numerical calculation.&quot;
Kerbiter5 months ago
Somehow the first time I see such pop up in my feed. Glad that someone (judging by the comments that is not the only one project) is working on this, of course I am rather far from the field but to me this feels like a step in the right direction for advancing AI past the hyperadvanced parrot stage that is the current &quot;AI&quot; is (at least per my perception).
rastierastie5 months ago
What do other HNers make out of this? Would you use this? Responsible for a legaltech startup here.
评论 #42547496 未加载
评论 #42548418 未加载
flimflamm5 months ago
Paper also here <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2409.13731" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;2409.13731</a>
评论 #42549858 未加载
dcreater5 months ago
Yet another RAG&#x2F;knowledge graph implementation.<p>At this point, the onus is on the developer to prove it&#x27;s value through AB comparisons versus traditional RAG. No person&#x2F;team has the bandwidth to try out this (n + 1) solution.
评论 #42546979 未加载
评论 #42550357 未加载
tessierashpool95 months ago
a quick look leaves me with the question:<p>what exactly is being tokenized? RDS, OWL, Neo4j, ...?<p>how is the knowledge graph serialized?
评论 #42548750 未加载
nextworddev4 months ago
Constructing and maintaining a knowledge graph is one thing.<p>Retrieving one with low latency is another.
ritiksharma234 months ago
How will this work on Scale??