Agentic Reasoning for Large Language Models
Paper
ā¢ 2601.12538 ā¢ Published
ā¢ 201
John Johnson
jjokah
AI & ML interests
Natural Language Processing, AI Engineering, Small Language Models.
Recent Activity
upvoted a paper 7 days ago
Agentic Reasoning for Large Language Models commented on their article 7 days ago
Small Language Models (SLM): A Comprehensive Overview commented on their article 7 days ago
Small Language Models (SLM): A Comprehensive Overview Organizations
upvoted a paper 7 days ago
commented on Small Language Models (SLM): A Comprehensive Overview 7 days ago
Here on Hugging Face (models section) use the Parameters filter and select the size range of the model you want.
commented on Small Language Models (SLM): A Comprehensive Overview 7 days ago
There has to be a trade-off here to really shrink the model size.
The model has to be scoped to specific tasks or domains rather than trying to be general-purpose.
And a good approach to achieve this is Knowledge Distillation: train a tiny student model on a specific task to mimic a larger teacher, but you lose generality. That's the trade-off.
upvoted an article 23 days ago
Article
One-Shot Any Web App with Gradio's gr.HTML
- +1
ā¢
25
upvoted an article about 1 month ago
Article
Introducing Daggr: Chain apps programmatically, inspect visually
-
- +3
ā¢
104
liked a model about 2 months ago
liked a Space about 2 months ago
liked a model about 2 months ago
upvoted 2 collections about 2 months ago
reacted to their post with š„ about 2 months ago
Post
1065
TranslateGemma: Open Translation Models (Jan 2026)
Google introduces TranslateGemma, a new suite of open translation models based on Gemma 3, available in 4B, 12B, and 27B parameter sizes.
Key Highlights:
ā¢ Supports 55 languages with high-quality translation across high-, mid-, and low-resource languages
ā¢ Exceptional efficiency: 12B model outperforms 27B baseline on WMT24++ benchmark
ā¢ Built using two-stage fine-tuning process distilling knowledge from Gemini models
ā¢ Retains strong multimodal capabilities (can translate text within images)
ā¢ Trained on nearly 500 additional language pairs for research adaptation
ā¢ Designed for diverse deployment environments from mobile to cloud
The models achieve state-of-the-art performance while maintaining exceptional efficiency, making high-quality translation accessible across different devices and use cases.
https://huggingface.co/collections/google/translategemma
Google introduces TranslateGemma, a new suite of open translation models based on Gemma 3, available in 4B, 12B, and 27B parameter sizes.
Key Highlights:
ā¢ Supports 55 languages with high-quality translation across high-, mid-, and low-resource languages
ā¢ Exceptional efficiency: 12B model outperforms 27B baseline on WMT24++ benchmark
ā¢ Built using two-stage fine-tuning process distilling knowledge from Gemini models
ā¢ Retains strong multimodal capabilities (can translate text within images)
ā¢ Trained on nearly 500 additional language pairs for research adaptation
ā¢ Designed for diverse deployment environments from mobile to cloud
The models achieve state-of-the-art performance while maintaining exceptional efficiency, making high-quality translation accessible across different devices and use cases.
https://huggingface.co/collections/google/translategemma
posted an
update about 2 months ago
Post
1065
TranslateGemma: Open Translation Models (Jan 2026)
Google introduces TranslateGemma, a new suite of open translation models based on Gemma 3, available in 4B, 12B, and 27B parameter sizes.
Key Highlights:
ā¢ Supports 55 languages with high-quality translation across high-, mid-, and low-resource languages
ā¢ Exceptional efficiency: 12B model outperforms 27B baseline on WMT24++ benchmark
ā¢ Built using two-stage fine-tuning process distilling knowledge from Gemini models
ā¢ Retains strong multimodal capabilities (can translate text within images)
ā¢ Trained on nearly 500 additional language pairs for research adaptation
ā¢ Designed for diverse deployment environments from mobile to cloud
The models achieve state-of-the-art performance while maintaining exceptional efficiency, making high-quality translation accessible across different devices and use cases.
https://huggingface.co/collections/google/translategemma
Google introduces TranslateGemma, a new suite of open translation models based on Gemma 3, available in 4B, 12B, and 27B parameter sizes.
Key Highlights:
ā¢ Supports 55 languages with high-quality translation across high-, mid-, and low-resource languages
ā¢ Exceptional efficiency: 12B model outperforms 27B baseline on WMT24++ benchmark
ā¢ Built using two-stage fine-tuning process distilling knowledge from Gemini models
ā¢ Retains strong multimodal capabilities (can translate text within images)
ā¢ Trained on nearly 500 additional language pairs for research adaptation
ā¢ Designed for diverse deployment environments from mobile to cloud
The models achieve state-of-the-art performance while maintaining exceptional efficiency, making high-quality translation accessible across different devices and use cases.
https://huggingface.co/collections/google/translategemma
upvoted 2 collections 3 months ago
reacted to their post with š 3 months ago
Post
274
LLM divide-and-conquer
w/ Massively Decomposed Agentic Processes
Ref: https://arxiv.org/abs/2511.09030
w/ Massively Decomposed Agentic Processes
Ref: https://arxiv.org/abs/2511.09030
posted an
update 4 months ago
Post
274
LLM divide-and-conquer
w/ Massively Decomposed Agentic Processes
Ref: https://arxiv.org/abs/2511.09030
w/ Massively Decomposed Agentic Processes
Ref: https://arxiv.org/abs/2511.09030
liked a Space 5 months ago
liked a model 6 months ago
liked 2 Spaces 6 months ago