LeapLab@THU retweeted
Microsoft releases ART
Anonymous Region Transformer for Variable Multi-Layer Transparent Image Generation
3
29
143
18,809
🚀 Excited to share our work on #ECCV2024: "AdaNAT: Exploring Adaptive Policy for Token-Based Image Generation".
🖼️ We introduce AdaNAT, a novel approach for efficient and high-quality image generation using adaptive policies in Non-autoregressive Transformers.
2
1
5
505
🔬 Read the paper: arxiv.org/abs/2409.00342
💻 Code & pretrained models: github.com/LeapLabTHU/AdaNAT
2
173
🔑 Key features:
Learnable policy network for adaptive modulation of token generation
Adversarial reward model for improved quality and diversity
Significantly reduced inference time compared to diffusion models
📊 Impressive results on ImageNet, MSCOCO, and CC3M datasets!
1
2
239
LeapLab@THU retweeted
very fantastic work done with @ALucyBrilliant @RayLu_THU @AndrewZ45732491. direct parameter editing can modulate LLM’s behavior
📢 New paper alert! 📢
🧠💉"Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing"
🚀 Modulate LLM behaviors through direct parameter editing.
🚀 Achieve up to 90% detoxification with inference-level computational cost! 💡🤖
arXiv: arxiv.org/abs/2407.08770
1
1
606
LeapLab@THU retweeted
ConvLLaVA
Hierarchical Backbones as Visual Encoder for Large Multimodal Models
High-resolution Large Multimodal Models (LMMs) encounter the challenges of excessive visual tokens and quadratic visual complexity. Current high-resolution LMMs address the quadratic
1
24
111
16,969
LeapLab@THU retweeted
📢Excited to share our recent work on Large Multimodal Models: ConvLLaVA. Without the encoding multiple image patches and multiple encoders, we use a hierarchical backbone, ConvNeXt, realizing high resolution understanding.
arxiv.org/pdf/2405.15738
1
1
3
290
LeapLab@THU retweeted
EfficientTrain++ is accepted by TPAMI2024🤩
🔥An off-the-shelf, easy-to-implement algorithm for training foundation visual backbones efficiently!
🔥1.5−3.0× lossless training/pre-training speedup on ImageNet-1K/22K!
Paper&Code:
arxiv.org/abs/2405.08768
github.com/LeapLabTHU/Effici…
4
7
942
Our recent work: Agent Attention!
[High Performance & Linear Complexity]
[Double the speed of SD and enhance generation quality, no additional fine-tuning is required]
Paper and code:
huggingface.co/papers/2312.0…
arxiv.org/abs/2312.08874
github.com/LeapLabTHU/Agent-…
2
20
77
18,276
Excited to share our #NeurIPS2023 spotlight paper! 🌟 It proposes a novel offline-to-online RL algorithm, efficiently utilizing collected samples by training a family of policies offline and selecting suitable ones online. Check out our paper for details! arxiv.org/abs/2310.17966
4
7
569
LeapLab@THU retweeted
ExpeL is now accepted at #AAAI24 ! The code and camera ready version will be updated promptly. Thanks for all the collaborators and see you in Vancouver!
🚀🚀🚀Our new preprint introduces ExpeL, an agent designed for self-improvement in LLM agents🤖. ExpeL can automatically gather experience, extract insights, and recall successful instances across tasks🧠. Checkout our paper at arxiv.org/abs/2308.10144.
1
3
5
756
LeapLab@THU retweeted
Check us out at #NeurIPS2023 poster！We investigate into Q-value divergence phenomenon in offline RL and find self-excitation to be the main reason. Using layernorm in RL models can fundamentally prevent this from happening. arxiv.org/pdf/2310.04411.pdf
1
4
8
541
Our recent work: Agent Attention!
[High Performance & Linear Complexity]
[Accelerate and improve Stable Diffusion, no additional fine-tuning is required]
The paper and code have been released:
arxiv.org/abs/2312.08874
github.com/LeapLabTHU/Agent-…
3
8
348