demirbey05

7 ms·

31.
by demirbey05 1y ago
Seems LLMs really hit the wall.
32.
by demirbey05 1y ago
That's what I meant. I don't think boss wants you to pay same money with less work time.
33.
by demirbey05 1y ago
> If an AI can replace these repeated tasks, I could spend more time with my fiancé, family, friends, and dog, which is awesome, and I am looking forward to that. I could not understand this optimism, aren't we living in a capitalis
34.
by demirbey05 1y ago
This is bad use of AI, we spend our compute to make science faster. I am pretty confident computational cost of this will be maybe 100x of chatgpt query. I don't want to think even environmental effects.
35.
by demirbey05 1y ago
We didn't even reach AGI and no sign with LLMs.
36.
by demirbey05 1y ago
>If AI soon becomes good enough at building software on its own, software engineering as we know it is dead. I have no interest in becoming a glorified project manager, orchestrating AI agents all day long. If it does happen, I am now co
37.
by demirbey05 1y ago
Appreciated > I will not be commenting on any self-reported AI competition performance results for which the methodology was not disclosed in advance of the competition.
38.
by demirbey05 1y ago
Google also joined IMO, and got gold prize. https://x.com/natolambert/status/1946569475396120653 OAI announced early, probably we will hear announcement from Google soon.
39.
by demirbey05 1y ago
I think from Canada team someone solved it but among all, its very few
40.
by demirbey05 1y ago
I mean progress speed, few months ago they released o3 it has 16 pt in imo 2025
41.
by demirbey05 1y ago
Progress is astounding. Recently report published about evaluation of LLMs on IMO 2025. o3 high didn't even get bronze. https://matharena.ai/imo/ Waiting for Terry Tao's thoughts, but these kind of things are
42.
Transformers Tutorial(github.com)
1 points by demirbey05 1y ago|0 comments
43.
by demirbey05 1y ago
I was really shocked when I first see this but yes it's in training data. Not thinking feature.
44.
by demirbey05 1y ago
How will you figure out or solve hallucinated assembly code ?
45.
PyTorch Reshaping with None(blog.detorch.xyz)
18 points by demirbey05 1y ago|0 comments
46.
Relic: Evaluating Compositional Instruction Following via Language Recognition(arxiv.org)
1 points by demirbey05 1y ago|0 comments
47.
by demirbey05 1y ago
omg, my ai agent did nil dereferencing, it seems it's trying to implement backdoor to my system so that it will crash my server.
48.
For Some Recent Graduates, the A.I. Job Apocalypse May Already Be Here(nytimes.com)
13 points by demirbey05 1y ago|23 comments
49.
by demirbey05 1y ago
If you invent someting better, you can get reward, lots of examples out there.
50.
by demirbey05 1y ago
https://stanford-cs336.github.io/spring2025/ Just learn this and its prerequisites
51.
by demirbey05 1y ago
Because its trained on human data.
52.
Show HN: AI app that turns educational videos into articles and quizzes(podco.xyz)
2 points by demirbey05 1y ago|0 comments
53.
InfiniteICL: Breaking the Limit of Context Window Size(arxiv.org)
1 points by demirbey05 2y ago|0 comments
54.
Operating Systems: A Linux Kernel-Oriented Perspective(cse.iitd.ac.in)
4 points by demirbey05 2y ago|0 comments
55.
The Future of Math with O1 Reasoning with Terence Tao(youtube.com)
2 points by demirbey05 2y ago|0 comments
56.
by demirbey05 2y ago
o1 is not included, I think each benchmark should include o1 and reasoning models. o-series is really changed the game.
57.
by demirbey05 2y ago
Think I mean Catan :)
58.
by demirbey05 2y ago
Yes its reinforcement learning, but need to create policy and each policy is specialized for specific tasks.
59.
by demirbey05 2y ago
I am not expert in llm reasoning but I think because of RL. You cannot use AlphaZero to play other games.
60.
by demirbey05 2y ago
its not like sonnet, yes current ai tools are increasing productivity and provides many ways to have chance to be rich, but agi is completely different. You need to handle evil competition between you and big fishes, probably big fishes wil

More ›