Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning
Paper
β’
2506.06205
β’
Published
β’
27
Junlin Zhou
jlzhou
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
3 days ago
Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal
Learning
commented on
a paper
3 days ago
Reinforcement Pre-Training
upvoted
a
paper
3 days ago
Reinforcement Pre-Training
Organizations
jlzhou's activity
upvoted
a
paper
3 days ago
commented
a
paper
3 days ago
upvoted
a
paper
3 days ago
upvoted
a
paper
17 days ago
liked
a
model
27 days ago
liked
a
model
about 1 month ago
upvoted
a
paper
about 1 month ago
liked
a
dataset
about 1 month ago
reacted to
mlabonne's
post with π
about 2 months ago
Post
14858
βοΈ AutoAbliteration
I made a Colab notebook to automatically abliterate models.
It's quite general, so you can do interesting stuff like blocking a given language in the model outputs.
π» Colab: https://colab.research.google.com/drive/1RmLv-pCMBBsQGXQIM8yF-OdCNyoylUR1?usp=sharing
I made a Colab notebook to automatically abliterate models.
It's quite general, so you can do interesting stuff like blocking a given language in the model outputs.
π» Colab: https://colab.research.google.com/drive/1RmLv-pCMBBsQGXQIM8yF-OdCNyoylUR1?usp=sharing
upvoted
an
article
about 2 months ago
Article
Uncensor any LLM with abliteration
By
β’
β’
610
liked
a
model
about 2 months ago
reacted to
etemiz's
post with π
about 2 months ago
Post
1619
Grok 3 Human Alignment Score: 42
It is better in health, nutrition, fasting compared to Grok 2. About the same in liberating tech like bitcoin and nostr. Worse in the misinformation and faith domains. The rest is about the same. So we have a model that is less faithful but knows how to live a healthier life.
https://sheet.zoho.com/sheet/open/mz41j09cc640a29ba47729fed784a263c1d08?sheetid=0&range=A1
https://huggingface.co/blog/etemiz/benchmarking-ai-human-alignment-of-grok-3
It is better in health, nutrition, fasting compared to Grok 2. About the same in liberating tech like bitcoin and nostr. Worse in the misinformation and faith domains. The rest is about the same. So we have a model that is less faithful but knows how to live a healthier life.
https://sheet.zoho.com/sheet/open/mz41j09cc640a29ba47729fed784a263c1d08?sheetid=0&range=A1
https://huggingface.co/blog/etemiz/benchmarking-ai-human-alignment-of-grok-3
upvoted
a
paper
about 2 months ago
liked
a
model
2 months ago
[bot] Conversion to Parquet
#1 opened 5 months ago
by
parquet-converter
[bot] Conversion to Parquet
#1 opened 5 months ago
by
parquet-converter
ε―δ»₯ζ·»ε δΈδΈLICENSEζδ»ΆεοΌ
2
#2 opened 3 months ago
by
adol-ch
upvoted
an
article
3 months ago
Article
You could have designed state of the art positional encoding
By
β’
β’
300
liked
a
model
3 months ago
reacted to
KaiChen1998's
post with π
3 months ago
Post
4887
π’ Our EMOVA paper has been accepted by CVPR 2025, and we are glad to release all resources, including code (training & inference), datasets (training & evaluation), and checkpoints (EMOVA-3B/7B/72B)!
π€ EMOVA is a novel end-to-end omni-modal LLM that can see, hear and speak. Given omni-modal (i.e., textual, visual and speech) inputs, EMOVA can generate both textual and speech responses with vivid emotional controls by utilizing the speech decoder and a style controller.
β¨ EMOVA Highlights
β State-of-the-art omni-modality: EMOVA achieves SoTA comparable results on both vision-language and speech benchmarks simultaneously.
β Device adaptation: our codebase supports training/inference on both NVIDIA GPUs (e.g., A800 & H20) and Ascend NPUs (e.g., 910B3)!
β Modular design: we integrate multiple implementations of vision encoder, vision projector, and language model, even including the most recent DeepSeekMoE-tiny!
π₯ You are all welcome to try and star!
- Project page: https://emova-ollm.github.io/
- Github: https://github.com/emova-ollm/EMOVA
- Demo: Emova-ollm/EMOVA-demo
π€ EMOVA is a novel end-to-end omni-modal LLM that can see, hear and speak. Given omni-modal (i.e., textual, visual and speech) inputs, EMOVA can generate both textual and speech responses with vivid emotional controls by utilizing the speech decoder and a style controller.
β¨ EMOVA Highlights
β State-of-the-art omni-modality: EMOVA achieves SoTA comparable results on both vision-language and speech benchmarks simultaneously.
β Device adaptation: our codebase supports training/inference on both NVIDIA GPUs (e.g., A800 & H20) and Ascend NPUs (e.g., 910B3)!
β Modular design: we integrate multiple implementations of vision encoder, vision projector, and language model, even including the most recent DeepSeekMoE-tiny!
π₯ You are all welcome to try and star!
- Project page: https://emova-ollm.github.io/
- Github: https://github.com/emova-ollm/EMOVA
- Demo: Emova-ollm/EMOVA-demo