thoughtlede

6 ms·

1.
by thoughtlede 5mo ago
Investor funds have been subsidizing the inference costs so far. Investors might move from funding the model providers to funding the enterprises that use those models. That is, they might move from funding the cost of the experiment to fun
2.
by thoughtlede 7mo ago
Strictly speaking, I don't think it is the generation or creation that diminishes their value. it is the consumption. You said it too: > If I see a million fake Tom Cruise videos, then it oversaturates my desire for desire for all T
3.
by thoughtlede 7mo ago
I think we can simplify the answer to this question for most audience and say "the air is blue". If they say, the air appears to be clear when I stare at something other than sky, the answer is you need more of air to be able to s
4.
by thoughtlede 10mo ago
For me there are two things about collaboration. Decision making is one, which you emphasized. The other is knowing what the collaboration brings to the table and shaping the rules of engagement to fit that expectation. Sometimes you collab
5.
by thoughtlede 1y ago
It boils down to whether your LLMs can speak graph queries better than SQL, for your use cases and data. As your data posture changes and your use cases change, you routinely reevaluate which DB query language suits best for LLMs. I'd
6.
by thoughtlede 1y ago
memorization + application = comprehension. Rinse and repeat. Whether leet code or anything else.
7.
by thoughtlede 1y ago
It feels to me that the hypothesis of this research was somewhat "begging the question". Reasoning models are trained to spit some tokens out that increase the chance of the models spitting the right answer at the end. That is, th
8.
by thoughtlede 2y ago
tadkar did a good job at explaining ColBERT. I understood ColBERT well in the context of where it lies on the spectrum of choices. On one side of the spectrum, you reduce each of the documents as well as the query to a lower-dimensional spa
9.
by thoughtlede 2y ago
When thinking about AI agents, there is still conflation between how to decide the next step to take vs what information is needed to decide the next step. If runtime information is insufficient, we can use AI/ML models to fill that in
10.
by thoughtlede 2y ago
> Beyond this, if you want to determine causality, e.g. whether events are "causally related" (happened before or after each other) or are "concurrent" (entirely independent of), you can look at Vector Clocks—I won&#x
11.
by thoughtlede 2y ago
Perhaps these findings might be indicating that we need more NN layers/attention blocks for performing reasoning. This project circumvented the lack of more trained layers by looping the input through currently trained layers more than
12.
by thoughtlede 2y ago
If function calling is sync, is MCP its async counterpart? Is that the gist of what MCP is? Open API (aka swagger) based function calling is standard already for sync calls, and it solves the NxM problem. I'm wondering if the proposed
13.
by thoughtlede 2y ago
I think the mention of 'spaghetti code' is a red herring from the author. If the output from an algorithm cannot be defined precisely as a function of the input, but you have some examples to show, that's where machine learni
14.
Classifier to detect DALL·E 3 images(openai.com)
2 points by thoughtlede 2y ago|0 comments
15.
by thoughtlede 2y ago
Interesting. I didn’t know this body of work. I haven’t read the documentation other than the abstract. If the protocol is about knowing what information to inject at what node in the network to achieve consensus, the protocol can (and will
16.
by thoughtlede 2y ago
That's interesting. In keyword-based indexing solutions, a document vector is created using "term frequency inverse document frequency" scores. The idea is to pump up the document on the dimension where the document is unique
17.
by thoughtlede 2y ago
LLMs are both language processing engines and knowledge bases. This article explores the knowledge base aspect of LLM and sheds light on the potential danger. The authors are well-justified in doing so because ChatGPT as a knowledge-bot is
18.
by thoughtlede 2y ago
Yes, the camera sees everything. But we could avoid teaching the model certain things. For instance, an undressed body and a dressed body could both be taught as a body. Likewise, medicine pills as well as regular mints could just be taught
19.
by thoughtlede 2y ago
Yeah. Rollbacks or reruns are hard when dealing with external systems. Actions need to be idempotent for reruns to work. One thing you may focus on is making workflows more durable: Checkpointing and sending to users summaries of last check
20.
by thoughtlede 2y ago
Thanks. What happens right now when the workflow fails mid-way? Do you ensure atomicity or durable execution?
21.
by thoughtlede 2y ago
Cool stuff. 1. For dynamic injection of arguments in your data plane, do you use LLMs? 2. What did you find you cannot do yet because of LLM limitations (and not because of lack of third-party integrations)? 3. I haven’t looked closely into
22.
by thoughtlede 3y ago
Answering to your second part of the question about hidden challenges: If you are using AI agents to automate a workflow [1] execution, then the question to ask is where is non-determinism in the workflow. As in, where do humans scratch the
23.
by thoughtlede 3y ago
Evolution of technology makes things unstable, not specifically servers. A decentralized application is not more stable than a centralized one. It depends on what's prioritized about the product. For example, you can still use SMTP ser
24.
by thoughtlede 3y ago
I have studied CRDTs at a deeper level for a few weeks and implemented several small prototypes. They are fascinating. As an eventual consistency model for data management, CRDT inspired techniques (op-based or state-based) are useful. Howe
25.
by thoughtlede 3y ago
In a multi-replica system, where, say, we cannot tolerate any failures or lags, is 2PC used in practice to achieve consensus? Or are there other methods for achieving such strict consensus?
26.
by thoughtlede 3y ago
Two-phase locking is different from two-phase commit, in spite of an overlap in their naming. Two-phase commit is relevant to be compared against Paxos - both of which fall under the category of consensus protocols. Two-phase locking is a c
27.
by thoughtlede 3y ago
I would characterize an LLM Agent to be "a software application of LLM" that is typically used to solve a problem that requires multiple steps. The application uses an LLM to breakdown the problem into pieces and asks the LLM to c
28.
by thoughtlede 3y ago
Part 1: https://www.computer.org/csdl/magazine/so/2023/01/09994260/1... (10.1109/MS.2022.3212165) Part 2: https://www.computer.org/csdl/magazine/so/2023&#x2F
29.
by thoughtlede 3y ago
I have implemented a JSON-like CRDT using this nice tutorial [1], which is a tree, but never an append-only DAG. I'm curious how two edges meeting the same vertex will make it difficult to resolve the conflict automatically? Is it that
30.
by thoughtlede 3y ago
> CRDTs don't _ensure_ ops are commutative (and associative, and idempotent) but rather they _require_ that ops are commutative (and associative, and idempotent) I disagree. You can create a CRDT flavor of data structure whose ops a

More ›