I saw that this project on GitHub has passed some tests using Kaggle datasets and won 22 gold medals. Could there be a problem of data contamination here? Can we try real-world competition problems on Kaggle?
NJX-njx
NJX-njx
AI & ML interests
AI infra, large model architecture, intelligent agent,evaluation
Recent Activity
View all activity
Organizations
I roughly get it, it's very interesting. It feels like this is an area that everyone hasn't delved into deeply so far.
I'm still very curious about what exactly these real-time data consist of.
replied to
marksverdhei's
post
5 days ago
Perhaps we can look forward to diffusion models
replied to
marksverdhei's
post
5 days ago
Perhaps, instead of thinking about architectural innovation, it is better to study how to reduce the cost of inference.
To be honest, hf has significant issues with functions such as payment and subscription, which greatly affect its user experience and user retention.
Your work is very meaningful.
I might not have fully understood. Do you deploy an agent to the GPU that you want to detect, and then use the agent to detect problems that the GPU may encounter during operation?
replied to
danielhanchen's
post
7 days ago
I want to know what this is provided to our model as. Is it a skill?
Actually, I think a very important point is that most independent developers do not have enough case studies to support their work, and at the same time, the cost of online deployment is actually a bit high
replied to
RakshitAralimatti's
post
16 days ago
To be honest, although it still looks like AI at first glance in terms of design and other aspects, the quality of the model's website has indeed improved a lot compared to previous models.
replied to
tegridydev's
post
16 days ago
So you want to create a dataset of papers that includes various AI-related papers, is that right?
replied to
rodrigomalossi's
post
16 days ago
I'm a bit confused. What can such synthetic census data be used for?
replied to
unmodeled-tyler's
post
17 days ago
Yep! You've got that exactly right. I've finetuned the model so that it when it responds, it's like a Type 1 personality from Enneagram.
great
posted
an
update
17 days ago
Post
177
Friends of the community, I have recently had some new ideas.
Some time ago, I came across a research analysis from two investors at a16z. In the past year of 2025, ChatGPT actually tried to promote some new AI functions in fields such as shopping, but in fact, the effect was not good.
I think the fundamental reason lies in the user's mindset, or rather, the user's interaction logic in vertical fields. The most prominent and distinctive feature of ChatGPT is that all-encompassing dialogue box, which is also a common problem with many homogeneous AI products nowadays (it seems that without a dialogue box, the AI's capabilities are sealed off).Although it can be adapted to many scenario fields, it will appear very boring in more vertical scenarios
Ask yourself, would you prefer the image-text waterfall flow interaction in shopping scenarios like Xiaohongshu, or the monotonous search box of ChatGPT? The answer is actually obvious from the start.
For all vertical scenarios, the interaction logic was already very well-developed before the emergence of AI. The user experience brought by such interaction logic is definitely not something that a single dialogue box can replace.
And if we want to create a good AI product in a vertical field, we should think more about how to silently embed the powerful capabilities of AI into the original interaction, and continuously iterate to provide users with a better experience.@lilianweng@clem@AdinaY
Some time ago, I came across a research analysis from two investors at a16z. In the past year of 2025, ChatGPT actually tried to promote some new AI functions in fields such as shopping, but in fact, the effect was not good.
I think the fundamental reason lies in the user's mindset, or rather, the user's interaction logic in vertical fields. The most prominent and distinctive feature of ChatGPT is that all-encompassing dialogue box, which is also a common problem with many homogeneous AI products nowadays (it seems that without a dialogue box, the AI's capabilities are sealed off).Although it can be adapted to many scenario fields, it will appear very boring in more vertical scenarios
Ask yourself, would you prefer the image-text waterfall flow interaction in shopping scenarios like Xiaohongshu, or the monotonous search box of ChatGPT? The answer is actually obvious from the start.
For all vertical scenarios, the interaction logic was already very well-developed before the emergence of AI. The user experience brought by such interaction logic is definitely not something that a single dialogue box can replace.
And if we want to create a good AI product in a vertical field, we should think more about how to silently embed the powerful capabilities of AI into the original interaction, and continuously iterate to provide users with a better experience.@lilianweng@clem@AdinaY
- 3 replies
This is very thought-provoking for me.
replied to
unmodeled-tyler's
post
17 days ago
Can I understand that you have fine-tuned using the Ministral 3 3B Instruct 2512 architecture, so that its own language style, decision-making, etc., are more in line with the characteristics of Type 1 personality in the Enneagram?
I'm a bit curious about how a 4B model like this performs in scenarios involving long instructions and multi-tool calls.
reacted to
their
post with 🔥
18 days ago
Post
1543
Hello, my friends,
Recently, I developed a tool in the field of AI learning: when a user inputs any knowledge point, the intelligent agent can write animation storyboards based on that knowledge point, generate the corresponding animation code for the storyboards through a coding agent, and finally render it into a video using the Manim engine.
The overall effect is similar to the videos from 3blue1brown. I hope that through such a tool, everyone can freely learn through videos of the same quality as 3b1b's.
However, I have recently encountered a problem regarding the video content. It is difficult to position geometric figures, symbols, etc., in the correct positions in the video, that is, there is a problem with positioning. I tried extracting video frames after generating the video and submitting them to a VLM for review to identify visual issues, and continuously modifying the prompts to optimize the generation quality, but the results were not satisfactory.
I wonder if anyone has any good methods to solve this positioning problem in the video.
Here is the project link: https://github.com/NJX-njx/code2video#
Recently, I developed a tool in the field of AI learning: when a user inputs any knowledge point, the intelligent agent can write animation storyboards based on that knowledge point, generate the corresponding animation code for the storyboards through a coding agent, and finally render it into a video using the Manim engine.
The overall effect is similar to the videos from 3blue1brown. I hope that through such a tool, everyone can freely learn through videos of the same quality as 3b1b's.
However, I have recently encountered a problem regarding the video content. It is difficult to position geometric figures, symbols, etc., in the correct positions in the video, that is, there is a problem with positioning. I tried extracting video frames after generating the video and submitting them to a VLM for review to identify visual issues, and continuously modifying the prompts to optimize the generation quality, but the results were not satisfactory.
I wonder if anyone has any good methods to solve this positioning problem in the video.
Here is the project link: https://github.com/NJX-njx/code2video#
posted
an
update
19 days ago
Post
1543
Hello, my friends,
Recently, I developed a tool in the field of AI learning: when a user inputs any knowledge point, the intelligent agent can write animation storyboards based on that knowledge point, generate the corresponding animation code for the storyboards through a coding agent, and finally render it into a video using the Manim engine.
The overall effect is similar to the videos from 3blue1brown. I hope that through such a tool, everyone can freely learn through videos of the same quality as 3b1b's.
However, I have recently encountered a problem regarding the video content. It is difficult to position geometric figures, symbols, etc., in the correct positions in the video, that is, there is a problem with positioning. I tried extracting video frames after generating the video and submitting them to a VLM for review to identify visual issues, and continuously modifying the prompts to optimize the generation quality, but the results were not satisfactory.
I wonder if anyone has any good methods to solve this positioning problem in the video.
Here is the project link: https://github.com/NJX-njx/code2video#
Recently, I developed a tool in the field of AI learning: when a user inputs any knowledge point, the intelligent agent can write animation storyboards based on that knowledge point, generate the corresponding animation code for the storyboards through a coding agent, and finally render it into a video using the Manim engine.
The overall effect is similar to the videos from 3blue1brown. I hope that through such a tool, everyone can freely learn through videos of the same quality as 3b1b's.
However, I have recently encountered a problem regarding the video content. It is difficult to position geometric figures, symbols, etc., in the correct positions in the video, that is, there is a problem with positioning. I tried extracting video frames after generating the video and submitting them to a VLM for review to identify visual issues, and continuously modifying the prompts to optimize the generation quality, but the results were not satisfactory.
I wonder if anyone has any good methods to solve this positioning problem in the video.
Here is the project link: https://github.com/NJX-njx/code2video#