blackcat201

5 ms·

1.
by blackcat201 9d ago
who says they aren't benchmaxxing?
2.
Joint Optimization of Tool Creation and Use for Large Language Model Agents(tool-use-smith.github.io)
5 points by blackcat201 22d ago|0 comments
3.
Training LLMs to write tools generalized beyond self use(arxiv.org)
6 points by blackcat201 22d ago|0 comments
4.
Challenge GPT and Claude to Run Their Own Lemonade Stands[video](youtube.com)
2 points by blackcat201 2mo ago|0 comments
5.
by blackcat201 2mo ago
I think the author need to rethink out of the box what's LLM routers in a traditional sense ( input in, route, output ) and move to think how a router would work in agentic workflow. See cognition Devin Fusion design.
6.
Computational Arbitrage in AI Model Markets via Routing, Finetuning(arxiv.org)
2 points by blackcat201 2mo ago|0 comments
7.
LLMs Refuse High-Cost Attacks but Stay Vulnerable to Cheap, Real-World Harm(expectedharm.github.io)
3 points by blackcat201 7mo ago|0 comments
8.
Open-Source Alternative to Claude Cowork Desktop App(github.com)
1 points by blackcat201 8mo ago|0 comments
9.
Kimi Linear: An Expressive, Efficient Attention Architecture(github.com)
217 points by blackcat201 11mo ago|47 comments
10.
Testing Image Generation Models with Explainable Human Evaluation(tiger-ai-lab.github.io)
2 points by blackcat201 11mo ago|0 comments
11.
What Chess (Might) Taught Us About Programming with AI(theblackcat102.github.io)
2 points by blackcat201 1y ago|0 comments
12.
by blackcat201 2y ago
Shameless plug, for anyone who's interested in "self-improvement" agent check out StreamBench[1] where we benchmark and try out what's essential for improvements in online settings. Basically we find feedback signal is v
13.
A Study on the Impact of Structured Output on Performance of LLMs(arxiv.org)
2 points by blackcat201 2y ago|0 comments
14.
by blackcat201 2y ago
Do beware on some reasoning task, our recent work[0] actually found it may cause some performance degradation as well as possible reasoning weakening in JSON. I really hope they fix this in the latest GPT-4o version. [0] https://
15.
Ambidex robot arms by Naver[video](youtube.com)
1 points by blackcat201 2y ago|0 comments
16.
by blackcat201 2y ago
The standard operation is to stop and check if any machine was out of calibration. So yes
17.
by blackcat201 3y ago
I own my LLM not because I need it now but having the luxury to fall back if openai ran out of money
18.
by blackcat201 3y ago
I have been following the vector database trend back in 2020 and I ended up with the conclusion: vector search features are a nice to have features which adds more value on existing database (postgres) or text search services (elasticsearch
19.
Why robotics companies fails (2021)[pdf](freshconsulting.com)
4 points by blackcat201 3y ago|0 comments
20.
by blackcat201 3y ago
https://theblackcat102.github.io/ Recently I am ranting the AI trends and some short writeup of things I read
21.
by blackcat201 3y ago
This looks pretty interesting! But the landing page has only one sentence : Understand and implement research papers faster, followed by a get in touch button. Care to extrapolate more? The blog button doesn't work as well?
22.
Unpopular Opinions on AI(theblackcat102.github.io)
2 points by blackcat201 3y ago|0 comments
23.
Chinese Traditional 3B Chat Model(huggingface.co)
1 points by blackcat201 3y ago|0 comments
24.
Short writeup on challenges of LLMs in May 2023(theblackcat102.github.io)
3 points by blackcat201 3y ago|0 comments
25.
by blackcat201 3y ago
Note that stability have been funding freelance researcher by providing compute resource such as RWKV[1], Open Assistant, some works by LAION[2] and lucidrains[3] [1] https://github.com/BlinkDL/RWKV-LM [2] https:&#x2F
26.
by blackcat201 4y ago
On the other hand, meta AI research should be renamed as OpenAI. They are the only few big institute who open almost every models they train (galactica, OPT, M2M, wav2vec ...)
27.
Intel Q4 Financial Report [pdf](d1io3yog0oux5.cloudfront.net)
2 points by blackcat201 4y ago|0 comments
28.
by blackcat201 4y ago
Recently just migrate a project from pypoetry away to the traditional setup method. Poetry works great for simple package, but once you started to add in complexities, it just falls apart due to everything was abstract away and simplified i
29.
by blackcat201 4y ago
When chatGPT first came out, my first thought was to replicate it myself. But then, I have too many missing skills or lack the time for backend, frontend and deployment. So I found LAION started an initiative for open-assistant. https:&#x
30.
Why Visa and Mastercard have yet to face their Kodak moment(ft.com)
2 points by blackcat201 4y ago|3 comments

More ›