HN user
17 USC 106 gives copyright holders exclusive rights to reproduce and distribute copies; no exemption exists for downloading digital copies because you own the physical book, and fair use (17 USC 107) is unlikely to apply when commercial alternatives exist and you’re copying entire works from unauthorized distributors.
Your observations for using a vector DB for retrieval-augmented generation are consistent with my own.
For my applications, I use pgvector since I can also use fulltext indexes and JOINs with the rest of my business logic which is stored in a postgres database. This also makes it easier to implement hybrid search, where the fulltext results and semantic search results are combined and reranked.
I think the main selling-point for standalone vector databases is scale, i.e., when you have a single "corpus" of over 10^7 chunks and embedding vectors that needs to serve hundreds of req/s. In my opinion, the overhead of maintaining a separate database that requires syncing with your primary database did not make sense for my application.
It's an incentives problem. At research universities, promotion is contingent on research output, and teaching is often seen as a distraction. At so-called teaching universities, promotion (or even survival) is mainly contingent on throughput, and not measures of pedagogical outcome.
If you are a teaching faculty at a university, it is against your own interests to invest time to develop novel teaching materials. The exception might be writing textbooks, which can be monetized, but typically are a net-negative endeavor.
There's a tool I use called Petal https://www.petal.org/reference-manager. The free tier allows up to 1GB of PDFs, which I believe are processed by GROBID and chunked for LLM QA.
The feature I find most useful is the table automation which I use for literature review, since it lets me run the same QA prompts on a collection of documents all at once.
> VUDA only takes kernels in SPIR-V format. VUDA does not provide any support for compiling CUDA C kernels directly to SPIR-V (yet). However, it does not know or care how the SPIR-V source was created - may it be GLSL, HLSL, OpenCL.
So the answer is no, it can't be used with kernels that use cublas or cudnn, which excludes almost all ML use-cases.
Here's a really cool interactive visualization of the Hamiltonian Monte Carlo algorithm, where you can choose different target distributions and play around with the algorithm parameters: https://chi-feng.github.io/mcmc-demo/app.html?algorithm=Hami...
Source code: https://github.com/chi-feng/mcmc-demo/blob/master/algorithms...
It also has visualizations for other flavors of Hamiltonian Monte Carlo, including the No-U-Turn Sampler (NUTS) used by Stan.
On your first point, there's some really good results from this paper:
Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers https://arxiv.org/abs/2301.02111
Website with examples: https://valle-demo.github.io/
For your second question, Apple is already rolling out AI-narrated audiobooks. See: https://arstechnica.com/gadgets/2023/01/apple-rolls-out-ai-n...