Fangyuan Yu PRO

Ksgk-fy's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

commented 3 papers about 2 months ago

Adaptive Length Image Tokenization via Recurrent Allocation

Paper • 2411.02393 • Published Nov 4 • 12 •

1

Inference Optimal VLMs Need Only One Visual Token but Larger Models

Paper • 2411.03312 • Published Nov 5 • 6 •

1

Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration

Paper • 2410.18076 • Published Oct 23 • 4 •

2

commented 3 papers 2 months ago

Lightweight Neural App Control

Paper • 2410.17883 • Published Oct 23 • 9 •

2

Autoregressive Large Language Models are Computationally Universal

Paper • 2410.03170 • Published Oct 4 • 1 •

1

Agent-as-a-Judge: Evaluate Agents with Agents

Paper • 2410.10934 • Published Oct 14 • 18 •

2

commented 8 papers 3 months ago

Emergent properties with repeated examples

Paper • 2410.07041 • Published Oct 9 • 8 •

3

FAN: Fourier Analysis Networks

Paper • 2410.02675 • Published Oct 3 • 24 •

6

Selective Attention Improves Transformer

Paper • 2410.02703 • Published Oct 3 • 23 •

2

Intelligence at the Edge of Chaos

Paper • 2410.02536 • Published Oct 3 • 6 •

2

Can Models Learn Skill Composition from Examples?

Paper • 2409.19808 • Published Sep 29 • 8 •

2

DynaMo: In-Domain Dynamics Pretraining for Visuo-Motor Control

Paper • 2409.12192 • Published Sep 18 • 4 •

3

DynaMo: In-Domain Dynamics Pretraining for Visuo-Motor Control

Paper • 2409.12192 • Published Sep 18 • 4 •

3

Scaling Smart: Accelerating Large Language Model Pre-training with Small Model Initialization

Paper • 2409.12903 • Published Sep 19 • 21 •

5

commented a paper 4 months ago

Iterative Graph Alignment

Paper • 2408.16667 • Published Aug 29 • 2 •

2

New activity in meta-llama/Llama-3.1-8B-Instruct 5 months ago

Tokenizer 'apply_chat_template' issue

1

#42 opened 5 months ago by

Ksgk-fy

what is the right tokenizer should I use for llama 3.1 8B?

3

#19 opened 5 months ago by

calebl

New activity in qresearch/llama-3.1-8B-vision-378 5 months ago

Great model!

1

#1 opened 5 months ago by

Ksgk-fy

New activity in neuralmagic/Mistral-7B-Instruct-v0.3-GPTQ-4bit 7 months ago

Inference GPU Ram requirement >60GB

1

#1 opened 7 months ago by

Ksgk-fy

commented a paper 9 months ago

The Unreasonable Ineffectiveness of the Deeper Layers

Paper • 2403.17887 • Published Mar 26 • 78 •

14