google/embeddinggemma-2
Feature Extraction • 0.7B • Updated • 116k • • 1.94k
Mengqi Li
Kullpar
AI & ML interests
None yet
Recent Activity
liked a model about 18 hours ago
google/embeddinggemma-2 liked a model 15 days ago
Edge0/Audio8-ASR-Infinite upvoted a paper 3 months ago
Loop the Loopies!Organizations
liked a
model about 18 hours ago
liked a
model 15 days ago
upvoted a paper 3 months ago
liked a
model 3 months ago
liked a
dataset 4 months ago
upvoted an article 5 months ago
Article
Harness, Scaffold, and the AI Agent Terms Worth Getting Right
sergiopaniego, ariG23498
• • 152 reacted to sergiopaniego's post with 👍👀 5 months ago
Post
10043
Harness, Scaffold, Context Engineering, Agent... do you actually know what they mean?
We wrote an AI agent glossary and tried to make sense of it all with simple definitions and real examples
↓ go read it ↓
https://huggingface.co/blog/agent-glossary
We wrote an AI agent glossary and tried to make sense of it all with simple definitions and real examples
↓ go read it ↓
https://huggingface.co/blog/agent-glossary
- 2 replies
upvoted a paper 5 months ago
liked a
model 5 months ago
upvoted a collection 5 months ago
liked a
dataset 5 months ago
liked 2
models 5 months ago
liked a
model 7 months ago
reacted to evalstate's post with 🤗 8 months ago
Post
3757
Hugging Face MCP Server v0.3.2
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Replace model_search and dataset_search with combined hub_repo_search tool.
- Less distracting description for hf_doc_search
- model_search and dataset_search tool calls will still function (plan to remove next release).
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Replace model_search and dataset_search with combined hub_repo_search tool.
- Less distracting description for hf_doc_search
- model_search and dataset_search tool calls will still function (plan to remove next release).
- 4 replies
liked a
model 8 months ago
reacted to andito's post with 🔥 over 1 year ago
Post
4148
🧠👁️ Can AI visualize solutions?
Humans often solve visual problems by sketching ideas in our minds. What if Vision-Language Models (VLMs) could do something similar, not by generating full images, but by using internal “mental sketches”?
That’s the idea behind Mirage, a new framework that empowers VLMs to reason using latent visual tokens. Instead of just thinking in words, Mirage mixes in abstract visual representations that help the model solve complex tasks.
These aren't photorealistic images. They're compact, internal representations optimized purely to support reasoning.
🔧 Mirage is trained in two phases:
1) Grounding: It learns to produce latent tokens anchored in real images.
2) Refinement: The model drops the images and learns to generate visual tokens on its own.
📈 And yes, it works!
On challenging benchmarks like Visual Spatial Planning, Jigsaw puzzles, and Spatial Attention Tasks, Mirage clearly outperforms GPT-4o and other strong baselines.
Smart sketches > empty words.
By mimicking the way humans visualize solutions, Mirage gives AI a new kind of imagination, one that’s faster, more efficient, and more human-like.
Kudos to the teams at UMass Amherst and MIT behind this exciting work.
Check the paper: Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens (2506.17218)
Humans often solve visual problems by sketching ideas in our minds. What if Vision-Language Models (VLMs) could do something similar, not by generating full images, but by using internal “mental sketches”?
That’s the idea behind Mirage, a new framework that empowers VLMs to reason using latent visual tokens. Instead of just thinking in words, Mirage mixes in abstract visual representations that help the model solve complex tasks.
These aren't photorealistic images. They're compact, internal representations optimized purely to support reasoning.
🔧 Mirage is trained in two phases:
1) Grounding: It learns to produce latent tokens anchored in real images.
2) Refinement: The model drops the images and learns to generate visual tokens on its own.
📈 And yes, it works!
On challenging benchmarks like Visual Spatial Planning, Jigsaw puzzles, and Spatial Attention Tasks, Mirage clearly outperforms GPT-4o and other strong baselines.
Smart sketches > empty words.
By mimicking the way humans visualize solutions, Mirage gives AI a new kind of imagination, one that’s faster, more efficient, and more human-like.
Kudos to the teams at UMass Amherst and MIT behind this exciting work.
Check the paper: Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens (2506.17218)
- 4 replies
reacted to yeonseok-zeticai's post with 👍🔥 over 1 year ago
Post
3323
Hi everyone,
I’ve been running small language models (SLLMs) directly on smartphones — completely offline, with no cloud backend or server API calls.
I wanted to share:
1. ⚡ Tokens/sec performance across several SLLMs
2. 🤖 Observations on hardware utilization (where the workload actually runs)
3. 📏 Trade-offs between model size, latency, and feasibility for mobile apps
There are reports for below models
- QWEN3 0.6B
- NVIDIA/Nemotron QWEN 1.5B
- SimpleScaling S1
- TinyLlama
- Unsloth tuned Llama 3.2 1B
- Naver HyperClova 0.5B
📜Comparable Benchmark reports (no cloud, all on-device):
I’d really value your thoughts on:
- Creative ideas to further optimize inference under these hardware constraints
- Other compact LLMs worth testing on-device
- Experiences you’ve had trying to deploy LLMs at the edge
If there’s interest, I’m happy to share more details on the test setup, hardware specs, or the tooling we used for these comparisons.
Thanks for taking a look, and you can build your own through at "https://mlange.zetic.ai"!
I’ve been running small language models (SLLMs) directly on smartphones — completely offline, with no cloud backend or server API calls.
I wanted to share:
1. ⚡ Tokens/sec performance across several SLLMs
2. 🤖 Observations on hardware utilization (where the workload actually runs)
3. 📏 Trade-offs between model size, latency, and feasibility for mobile apps
There are reports for below models
- QWEN3 0.6B
- NVIDIA/Nemotron QWEN 1.5B
- SimpleScaling S1
- TinyLlama
- Unsloth tuned Llama 3.2 1B
- Naver HyperClova 0.5B
📜Comparable Benchmark reports (no cloud, all on-device):
I’d really value your thoughts on:
- Creative ideas to further optimize inference under these hardware constraints
- Other compact LLMs worth testing on-device
- Experiences you’ve had trying to deploy LLMs at the edge
If there’s interest, I’m happy to share more details on the test setup, hardware specs, or the tooling we used for these comparisons.
Thanks for taking a look, and you can build your own through at "https://mlange.zetic.ai"!