Seonil Son

sonsus's activity

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

upvoted 2 papers 5 days ago

SpatialBoost: Enhancing Visual Representation through Language-Guided Reasoning

Paper • 2603.22057 • Published 6 days ago • 44

RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models

Paper • 2603.21341 • Published 7 days ago • 23

updated a model 3 months ago

sonsus/gemma-3-0.7b-vlm-custom

0.7B • Updated Dec 24, 2025 • 2

published a model 3 months ago

sonsus/gemma-3-0.7b-vlm-custom

0.7B • Updated Dec 24, 2025 • 2

liked a model 3 months ago

GuminiResearch/Gumini-1.5B-Base

Text Generation • 2B • Updated Dec 18, 2025 • 75 • 8

liked a model 4 months ago

GuminiResearch/Gumini-1B-Base

Text Generation • 1B • Updated Dec 17, 2025 • 35 • 7

New activity in google/gemma-3-4b-pt 4 months ago

Wrong configs

❤️ 6

4

#5 opened about 1 year ago by

pere

New activity in NCSOFT/harim_plus 5 months ago

Create test.txt

#1 opened 5 months ago by

sonsus

upvoted a paper 6 months ago

VARCO-VISION-2.0 Technical Report

Paper • 2509.10105 • Published Sep 12, 2025 • 6

New activity in lerobot/act_aloha_sim_transfer_cube_human 6 months ago

Fix: hyperlink to `eval.py` script corrected.

#6 opened 6 months ago by

sonsus

liked a dataset 7 months ago

fgenie777/Arena-Lite-Experiments-Result-Data

Updated Aug 4, 2025 • 6 • 1

New activity in fgenie777/Arena-Lite-Experiments-Result-Data 8 months ago

Update README.md

#1 opened 8 months ago by

sonsus

updated a Space 8 months ago

Running

1

Arena-Lite

🔥

1

Tournaments for Efficient & Reliable LLM Benchmarking

New activity in NCSOFT/ArenaLite 8 months ago

Adding your own judge prompt: Working example

#2 opened 8 months ago by

sonsus

Submitting OPENAI API KEY is quite worrisome: Local-hosting of Arena-Lite

#1 opened 8 months ago by

sonsus

published a Space 8 months ago

Running

1

Arena-Lite

🔥

1

Tournaments for Efficient & Reliable LLM Benchmarking

updated a Space 8 months ago

Running

11

HaRiM+

🤗

11

HaRiM+ is a reference-free summary faithfulness measure.

upvoted a paper 8 months ago

Varco Arena: A Tournament Approach to Reference-Free Benchmarking Large Language Models

Paper • 2411.01281 • Published Nov 2, 2024 • 7

liked 2 models 9 months ago

NCSOFT/GME-VARCO-VISION-Embedding

Feature Extraction • 8B • Updated Jan 8 • 170 • 11

NCSOFT/VARCO-VISION-2.0-14B

Image-Text-to-Text • 15B • Updated Sep 15, 2025 • 365 • 45