How AI Impacts Skill Formation
Paper ā¢ 2601.20245 ā¢ Published ā¢ 10
nltpt-q
community
AI & ML interests
None defined yet.
cfahlgren1Ā
submitted a
paper to Daily Papers 3 months ago
Post
4413
š What happened in AI in 2025? š
We prepared the 2025 version of the HF AI Timeline Grid, highlighting open vs API-based model releases, and allowing you to browse and filter by access, modality, and release type!
Play with it here:
2025-ai-timeline/2025-ai-timeline
Here's my personal quarterly TL;DR:
1ļøā£ Q1 ā Learning to Reason
Deepseek not only releases a top-notch reasoning model, but shows how to train them and compete with closed frontier models. OpenAI debuts Deep Research.
Significant milestones: DeepSeek R1 & R1-Zero, Qwen 2.5 VL, OpenAI Deep Research, Gemini 2.5 Pro (experimental)
2ļøā£ Q2 ā Multimodality and Coding
More LLMs embrace multimodality by default, and there's a surge in coding agents. Strong vision, audio, and generative models emerge.
Significant milestones: Llama 4, Qwen 3, Imagen 4, OpenAI Codex, Google Jules, Claude 4
3ļøā£ Q3 ā "Gold" rush, OpenAI opens up, the community goes bananas
Flagship models get gold in Math olympiads and hard benchmarks. OpenAI releases strong open source models and Google releases the much anticipated nano-banana for image generation and editing. Agentic workflows become commonplace.
Significant milestones: Gemini and OpenAI IMO Gold, gpt-oss, Gemini 2.5 Flash Image, Grok 4, Claude Sonnet 4.5
4ļøā£ Q4 ā Mistral returns, leaderboard hill-climbing
Mistral is back with updated model families. All labs release impressive models to wrap up the year!
Significant milestones: Claude Opus 4.5, DeepSeek Math V2, FLUX 2, GPT 5.1, Kimi K2 Thinking, Nano Banana Pro, GLM 4.7, Gemini 3, Mistral 3, MiniMax M2.1 š¤Æ
Credits
š NHLOCAL for the source data https://github.com/NHLOCAL/AiTimeline
š«” @reach-vb for the original idea, design and recipe
š @ariG23498 and yours truly for compiling and verifying the 2025 edition
š„³ Here's to 2026, wishing it becomes the best year ever for open releases and on-device-first use-cases! š„
We prepared the 2025 version of the HF AI Timeline Grid, highlighting open vs API-based model releases, and allowing you to browse and filter by access, modality, and release type!
Play with it here:
2025-ai-timeline/2025-ai-timeline
Here's my personal quarterly TL;DR:
1ļøā£ Q1 ā Learning to Reason
Deepseek not only releases a top-notch reasoning model, but shows how to train them and compete with closed frontier models. OpenAI debuts Deep Research.
Significant milestones: DeepSeek R1 & R1-Zero, Qwen 2.5 VL, OpenAI Deep Research, Gemini 2.5 Pro (experimental)
2ļøā£ Q2 ā Multimodality and Coding
More LLMs embrace multimodality by default, and there's a surge in coding agents. Strong vision, audio, and generative models emerge.
Significant milestones: Llama 4, Qwen 3, Imagen 4, OpenAI Codex, Google Jules, Claude 4
3ļøā£ Q3 ā "Gold" rush, OpenAI opens up, the community goes bananas
Flagship models get gold in Math olympiads and hard benchmarks. OpenAI releases strong open source models and Google releases the much anticipated nano-banana for image generation and editing. Agentic workflows become commonplace.
Significant milestones: Gemini and OpenAI IMO Gold, gpt-oss, Gemini 2.5 Flash Image, Grok 4, Claude Sonnet 4.5
4ļøā£ Q4 ā Mistral returns, leaderboard hill-climbing
Mistral is back with updated model families. All labs release impressive models to wrap up the year!
Significant milestones: Claude Opus 4.5, DeepSeek Math V2, FLUX 2, GPT 5.1, Kimi K2 Thinking, Nano Banana Pro, GLM 4.7, Gemini 3, Mistral 3, MiniMax M2.1 š¤Æ
Credits
š NHLOCAL for the source data https://github.com/NHLOCAL/AiTimeline
š«” @reach-vb for the original idea, design and recipe
š @ariG23498 and yours truly for compiling and verifying the 2025 edition
š„³ Here's to 2026, wishing it becomes the best year ever for open releases and on-device-first use-cases! š„
- 3 replies
ariG23498Ā
authored a
paper 6 months ago
Post
2088
New post is live!
This time we cover some major updates to
š¤
This time we cover some major updates to
transformers.š¤
- 2 replies
Post
944
I have always advocated for writing techinical stories without using LLMs.
The following one page editorial really drives the point home.
https://www.nature.com/articles/s44222-025-00323-4
The following one page editorial really drives the point home.
https://www.nature.com/articles/s44222-025-00323-4
cfahlgren1Ā
posted an update 10 months ago
Post
1125
I ran the Anthropic Misalignment Framework for a few top models and added it to a dataset: cfahlgren1/anthropic-agentic-misalignment-results
You can read the reasoning traces of the models trying to blackmail the user and perform other actions. It's very interesting!!
You can read the reasoning traces of the models trying to blackmail the user and perform other actions. It's very interesting!!
cfahlgren1Ā
posted an update 11 months ago
Post
419
Really nice to see AllenAI drop the Reward-Bench-2 dataset and leaderboard from their new paper all on the hub! š
allenai/reward-bench
allenai/reward-bench-2
allenai/reward-bench-2-results
Great work @natolambert , allenai and others!! š¤
allenai/reward-bench
allenai/reward-bench-2
allenai/reward-bench-2-results
Great work @natolambert , allenai and others!! š¤
cfahlgren1Ā
posted an update 11 months ago
Post
1740
Yesterday, we dropped a new conversational viewer for datasets on the hub! š¬
Actually being able to view and inspect your data is extremely important. This is a big step in making data more accessible and actionable for everyone.
Here's some datasets you can try it out on:
ā¢ mlabonne/FineTome-100k
ā¢ Salesforce/APIGen-MT-5k
ā¢ open-thoughts/OpenThoughts2-1M
ā¢ allenai/tulu-3-sft-mixture
Any other good ones?
Actually being able to view and inspect your data is extremely important. This is a big step in making data more accessible and actionable for everyone.
Here's some datasets you can try it out on:
ā¢ mlabonne/FineTome-100k
ā¢ Salesforce/APIGen-MT-5k
ā¢ open-thoughts/OpenThoughts2-1M
ā¢ allenai/tulu-3-sft-mixture
Any other good ones?
- 1 reply
pcuenqĀ
authored a
paper about 1 year ago
cfahlgren1Ā
authored a
paper about 1 year ago
cfahlgren1Ā
posted an update about 1 year ago
Post
2357
If you haven't seen yet, we just released Inference Providers š
> 4 new serverless inference providers on the Hub š¤Æ
> Use your HF API key or personal key with all providers š
> Chat with Deepseek R1, V3, and more on HF Hub š
> We support Sambanova, TogetherAI, Replicate, and Fal.ai šŖ
Best of all, we don't charge any markup on top of the provider š«° Have you tried it out yet? HF Pro accounts get $2 of free usage for the provider inference.
> 4 new serverless inference providers on the Hub š¤Æ
> Use your HF API key or personal key with all providers š
> Chat with Deepseek R1, V3, and more on HF Hub š
> We support Sambanova, TogetherAI, Replicate, and Fal.ai šŖ
Best of all, we don't charge any markup on top of the provider š«° Have you tried it out yet? HF Pro accounts get $2 of free usage for the provider inference.
Post
2862
Tried my hand at simplifying the derivations of Direct Preference Optimization.
I cover how one can reformulate RLHF into DPO. The idea of implicit reward modeling is chef's kiss.
Blog: https://huggingface.co/blog/ariG23498/rlhf-to-dpo
I cover how one can reformulate RLHF into DPO. The idea of implicit reward modeling is chef's kiss.
Blog: https://huggingface.co/blog/ariG23498/rlhf-to-dpo
Post
2094
Timm ā¤ļø Transformers
Wtih the latest version of transformers you can now use any timm model with the familiar transformers API.
Blog Post: https://huggingface.co/blog/timm-transformers
Repository with examples: https://github.com/ariG23498/timm-wrapper-examples
Collection: ariG23498/timmwrapper-6777b85f1e8d085d3f1374a1
Wtih the latest version of transformers you can now use any timm model with the familiar transformers API.
Blog Post: https://huggingface.co/blog/timm-transformers
Repository with examples: https://github.com/ariG23498/timm-wrapper-examples
Collection: ariG23498/timmwrapper-6777b85f1e8d085d3f1374a1
cfahlgren1Ā
posted an update over 1 year ago
Post
1793
Wow, I just added Langfuse tracing to the Deepseek Artifacts app and it's really nice š„
It allows me to visualize and track more things along with the cfahlgren1/react-code-instructions dataset.
It was just added as a one click Docker Space template, so it's super easy to self host šŖ
It allows me to visualize and track more things along with the cfahlgren1/react-code-instructions dataset.
It was just added as a one click Docker Space template, so it's super easy to self host šŖ
cfahlgren1Ā
posted an update over 1 year ago
Post
2283
You'll notice the AI in the SQL Console is much better at working with chatml conversations:
Here's example of unnesting the cfahlgren1/react-code-instructions in less than 10 seconds by asking it. Check it out here: cfahlgren1/react-code-instructions
- "show me the average assistant response length"
- "extract user, system, and assistant messages into separate columns"
It's super easy to work with conversational datasets now with natural language š£ļø
Here's example of unnesting the cfahlgren1/react-code-instructions in less than 10 seconds by asking it. Check it out here: cfahlgren1/react-code-instructions
- "show me the average assistant response length"
- "extract user, system, and assistant messages into separate columns"
It's super easy to work with conversational datasets now with natural language š£ļø
- 2 replies
cfahlgren1Ā
posted an update over 1 year ago
Post
3356
The deepseek-ai/DeepSeek-V3 is very good! I have been playing with it and found it is really good at one-shotting a pretty good landing page.
You can play with it here: https://deepseek-artifacts.vercel.app
All the responses get saved in the cfahlgren1/react-code-instructions dataset. Hopefully we can build one of the biggest, highest quality frontend datasets on the hub šŖ
You can play with it here: https://deepseek-artifacts.vercel.app
All the responses get saved in the cfahlgren1/react-code-instructions dataset. Hopefully we can build one of the biggest, highest quality frontend datasets on the hub šŖ
Post
1473
We are blessed with another iteration of Pali Gemma. Google launches PaliGemma 2.
google/paligemma-2-release-67500e1e1dbfdd4dee27ba48
merve/paligemma2-vqav2
google/paligemma-2-release-67500e1e1dbfdd4dee27ba48
merve/paligemma2-vqav2
cfahlgren1Ā
posted an update over 1 year ago
Post
1948
You can just ask things š£ļø
"show me messages in the coding category that are in the top 10% of reward model scores"
Download really high quality instructions from the Llama3.1 405B synthetic dataset š„
argilla/magpie-ultra-v1.0
"show me messages in the coding category that are in the top 10% of reward model scores"
Download really high quality instructions from the Llama3.1 405B synthetic dataset š„
argilla/magpie-ultra-v1.0
cfahlgren1Ā
posted an update over 1 year ago
Post
3085
We just dropped an LLM inside the SQL Console š¤Æ
The amazing, new Qwen/Qwen2.5-Coder-32B-Instruct model can now write SQL for any Hugging Face dataset āØ
It's 2025, you shouldn't be hand writing SQL! This is a big step in making it where anyone can do in depth analysis on a dataset. Let us know what you think š¤
The amazing, new Qwen/Qwen2.5-Coder-32B-Instruct model can now write SQL for any Hugging Face dataset āØ
It's 2025, you shouldn't be hand writing SQL! This is a big step in making it where anyone can do in depth analysis on a dataset. Let us know what you think š¤