arroyo VS trulens

Interesting - I looked into your code a bit. I found your window aggregation library [1]. You may be interested in looking into the Rust implementation of some of the research work I've been a part of [2].

In Flink, I believe the reason they need to implement their own backpressure system is that they multiplex TCP connections. That is, they have multiple logical streams flowing through a single TCP connection. If that's the case, you need to do some work to 1) detect which logical stream is the one that's blocking, and 2) don't block because other logical streams may be able to use the active TCP connection.

Thinking it through, I think what Flink's approach buys is not necessarily better performance, but better just a manageable number of connections. That is, imagine you have a process P1 with operators A, B and C. And then P2 has D, E, F. Now imagine that this is a shuffle, where A, B and C are fully connected to D, E and F. In my old system, you would have 9 TCP connections. In Flink, you will have 1.

[1] https://github.com/ArroyoSystems/arroyo/blob/master/arroyo-w...

Retrieval using a single vector is called dense passage retrieval (DPR), because an entire passage (dozens to hundreds of tokens) is encoded as a single vector. ColBERT instead encodes a vector-per-token, where each vector is influenced by surrounding context. This leads to meaningfully better results; for example, here’s ColBERT running on Astra DB compared to DPR using openai-v3-small vectors, compared with TruLens for the Braintrust Coda Help Desk data set. ColBERT easily beats DPR at correctness, context relevance, and groundedness.

12. TruLens by TruEra | Github | tutorial

3) Hallucination is probably the biggest problem we solve for. To do evals for hallucination, we typically see our users use a combination of groundedness (does the context support the LLM response) and context relevance (is the retrieved context relevant to the query). There's also a bunch more for the evaluations you mentioned (moderation models, sentiment, usefulness, etc.) and it's pretty easy to add custom evals.

Also - my hot take is that gpt-3.5 is good enough for evals (sometimes better) than gpt-4 if you give the LLM enough instructions on how to do the eval.

website: https://www.trulens.org/

The team at TruEra recently released an open source project for evaluation & tracking of LLM applications called TruLens-Eval. We’ve specifically targeted retrieval-augmented QA as a core use case and so far we’ve seen it used for comparing different models and parameters, prompts, vector-db configurations and query planning strategies. I’d love to get your feedback on it.