Post
990
Kimi K2 is a bit disappointing by my expectations. It is on a par with Codex mini.
onekq-ai/WebApp1K-models-leaderboard
onekq-ai/WebApp1K-models-leaderboard
- 1 reply
Yi Cui
onekq
AI & ML interests
Benchmark, Code Generation Model
Recent Activity
replied to
their
post
1 day ago
Context rot is such a catchy phrase, but the problem has been identified 2+ years ago, called attention decay.
https://huggingface.co/papers/2307.03172
I spotted the same problem in coding tasks, and documented in my book (https://www.amazon.com/dp/9999331130).
Why did this problem become hot again? This is because many of us thought the problem has been solved by long context models, which is not true.
Here we were misled by benchmarks. Most long-context benchmarks build around the QA scenario, i.e. "finding needle in haystack". But in agentic scenarios, the model needs to find EVERYTHING in the haystack, and just can't afford enough attention for this challenge.
replied to
their
post
5 days ago
Context rot is such a catchy phrase, but the problem has been identified 2+ years ago, called attention decay.
https://huggingface.co/papers/2307.03172
I spotted the same problem in coding tasks, and documented in my book (https://www.amazon.com/dp/9999331130).
Why did this problem become hot again? This is because many of us thought the problem has been solved by long context models, which is not true.
Here we were misled by benchmarks. Most long-context benchmarks build around the QA scenario, i.e. "finding needle in haystack". But in agentic scenarios, the model needs to find EVERYTHING in the haystack, and just can't afford enough attention for this challenge.
Organizations
posted
an
update
1 day ago
replied to
their
post
1 day ago
I checked Codex and Claude, and they both use their own models to compress.
For OCR, as you stated lots of training would need to be done. It's an ecosystem problem.
replied to
their
post
5 days ago
Good stuff! I didn't consider token cost at all.
I'm thinking about an open source project for a context compressor (algorithmic, at most a small on-premise model) for agent builders. Does this make sense? If so, how should it look like?
replied to
their
post
5 days ago
Do you know any work that studied how agents use context?
posted
an
update
5 days ago
Post
3597
Context rot is such a catchy phrase, but the problem has been identified 2+ years ago, called attention decay.
Lost in the Middle: How Language Models Use Long Contexts (2307.03172)
I spotted the same problem in coding tasks, and documented in my book (https://www.amazon.com/dp/9999331130).
Why did this problem become hot again? This is because many of us thought the problem has been solved by long context models, which is not true.
Here we were misled by benchmarks. Most long-context benchmarks build around the QA scenario, i.e. "finding needle in haystack". But in agentic scenarios, the model needs to find EVERYTHING in the haystack, and just can't afford enough attention for this challenge.
Lost in the Middle: How Language Models Use Long Contexts (2307.03172)
I spotted the same problem in coding tasks, and documented in my book (https://www.amazon.com/dp/9999331130).
Why did this problem become hot again? This is because many of us thought the problem has been solved by long context models, which is not true.
Here we were misled by benchmarks. Most long-context benchmarks build around the QA scenario, i.e. "finding needle in haystack". But in agentic scenarios, the model needs to find EVERYTHING in the haystack, and just can't afford enough attention for this challenge.
-
- 6 replies
replied to
their
post
6 days ago
That is the case.
I'm a developer at heart. As a developer, majority of your time is spent running things and hop around environments, e.g. IDE, Cloud, Github. These environments all happen to have full featured bash support, a perfect sandbox for the CLI form factor.
The paradigm change AI brought to the developer world is nothing short of meteoric, but also an exception. Lots of efforts try to generalize the momentum to the next area(s). I won't bet on them.
posted
an
update
6 days ago
Post
4149
I am on the model layer and focus on atomic tasks, so I don't get involved in product discussions. But this provocative article provoked the community quite a bit. The case in point is Claude Code, which happens to be my biggest productivity revolution since ChatGPT.
RAG predated TUI and agents. So to be fair it's quite an achievement to survive the AI evolution. But I feel it is overshadowed by context engineering in the agent era. How does everyone feel about this?
https://www.nicolasbustamante.com/p/the-rag-obituary-killed-by-agents
RAG predated TUI and agents. So to be fair it's quite an achievement to survive the AI evolution. But I feel it is overshadowed by context engineering in the agent era. How does everyone feel about this?
https://www.nicolasbustamante.com/p/the-rag-obituary-killed-by-agents
-
- 2 replies
posted
an
update
9 days ago
Post
314
I tried to test the DeepSeek OCR model on a diagram-to-SQL task, i.e. visualize SQL schema to a E-R diagram, the combine the diagram with natural language question as prompt. The model outputs SQL query but unusable. The multimodality model (DeepSeek VL) performs better, but the good old coding LLM is far better.
So this model is still, and meant to be, an OCR model. It does compress long context in a new way, but will have to be trained for other tasks will long context will be applied. OCR itself doesn't need long context.
TLDR: lots of work will have to be done to make this main stream.
So this model is still, and meant to be, an OCR model. It does compress long context in a new way, but will have to be trained for other tasks will long context will be applied. OCR itself doesn't need long context.
TLDR: lots of work will have to be done to make this main stream.
posted
an
update
10 days ago
Post
270
I wrote this article to explain the difference between vision token and text token. They are apples and oranges, but also the source of compression efficiency of DeepSeek OCR (don't forget Glyph by THUDM!)
https://huggingface.co/blog/onekq/behind-each-token
I am running experiment with DeepSeek OCR BTW
https://huggingface.co/blog/onekq/behind-each-token
I am running experiment with DeepSeek OCR BTW
posted
an
update
11 days ago
Post
172
Random thought on 🐋DeepSeek🐋 OCR model: layout software/design will be hot, presentation matters way more than wording.
posted
an
update
18 days ago
Post
223
DeepSeek 3.2-exp is doing better than v3, but behind R1. This is quite interesting.
onekq-ai/WebApp1K-models-leaderboard
onekq-ai/WebApp1K-models-leaderboard
posted
an
update
30 days ago
Post
322
WebApp1K measures an oldest and simplest kind of task predated ChatGPT. It is code completion, you can also consider it a translation task mapping test spec into code. It requires no conversation, reasoning (which helps sometimes), or RL.
I don't think it is on the roadmap of top labs. Otherwise, you can't explain why Claude 4 has the same 70+ score on SweBench, which is way more challenging than this benchmark.
Neither do I encourage model builders to optimize towards my benchmark, which in itself won't be too hard to top the leaderboard. I just argue that we're still in a very early phase.
What I witness now is still the same pattern: the dropping of generic models strategically optimized towards famous benchmarks. Meanwhile, agent builders (top labs and startups alike) painfully prompt these models to follow their expectations, and pray they won't drift overnight.
I don't think it is on the roadmap of top labs. Otherwise, you can't explain why Claude 4 has the same 70+ score on SweBench, which is way more challenging than this benchmark.
Neither do I encourage model builders to optimize towards my benchmark, which in itself won't be too hard to top the leaderboard. I just argue that we're still in a very early phase.
What I witness now is still the same pattern: the dropping of generic models strategically optimized towards famous benchmarks. Meanwhile, agent builders (top labs and startups alike) painfully prompt these models to follow their expectations, and pray they won't drift overnight.
posted
an
update
about 1 month ago
Post
226
GPT OSS is as of now the top open source model, whose performance is very close to Claude and GPT-5, and above all other models.
onekq-ai/WebApp1K-models-leaderboard
onekq-ai/WebApp1K-models-leaderboard
posted
an
update
about 1 month ago
Post
275
My book on WebApp1K is published on Amazon
https://www.amazon.com/dp/9999331130
https://elivabooks.com/en/book/book-8002082768
https://www.amazon.com/dp/9999331130
https://elivabooks.com/en/book/book-8002082768
posted
an
update
about 1 month ago
Post
364
Lots of users don't know what WebApp1K tests. I published two datasets for the tasks and generations by models.
onekq-ai/WebApp1K-React-Generations
onekq-ai/WebApp1K-Duo-React-Generations
onekq-ai/WebApp1K-React-Generations
onekq-ai/WebApp1K-Duo-React-Generations
posted
an
update
about 1 month ago
Post
4419
Claude Opus 4.1 is slightly better than Opus 4, but still behind GPT-5
onekq-ai/WebApp1K-models-leaderboard
onekq-ai/WebApp1K-models-leaderboard
- 3 replies
replied to
their
post
about 1 month ago
is codex available on API? If so, we should test
posted
an
update
about 2 months ago
Post
170
GPT-5 achieves new SOTA! But I need to evaluate more models after a few months of gap.
onekq-ai/WebApp1K-models-leaderboard
onekq-ai/WebApp1K-models-leaderboard
-
- 2 replies