shawn

ares0728's activity

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

upvoted 2 papers 2 days ago

Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models

Paper • 2602.02185 • Published 3 days ago • 120

Vision-DeepResearch: Incentivizing DeepResearch Capability in Multimodal Large Language Models

Paper • 2601.22060 • Published 7 days ago • 142

upvoted a paper 8 days ago

AdaReasoner: Dynamic Tool Orchestration for Iterative Visual Reasoning

Paper • 2601.18631 • Published 10 days ago • 47

updated a dataset 3 months ago

ares0728/ares-rl-verl-format

Viewer • Updated Nov 3, 2025 • 82k • 26

published 2 datasets 3 months ago

ares0728/ares-rl-verl-format

Viewer • Updated Nov 3, 2025 • 82k • 26

ares0728/all-rl-data-sum

Updated Nov 3, 2025

upvoted a paper 4 months ago

ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping

Paper • 2510.08457 • Published Oct 9, 2025 • 13

updated a dataset 4 months ago

ares0728/ARES-Adaptive-Coldstart

Updated Oct 12, 2025 • 35

updated 2 models 4 months ago

ares0728/ARES-Coldstart-7B

Updated Oct 12, 2025

ares0728/ARES-RL-7B

Updated Oct 12, 2025 • 2

updated a dataset 4 months ago

ares0728/ARES-hard-validation

Viewer • Updated Oct 12, 2025 • 2.46k • 11

updated a collection 4 months ago

ARES

Collection

🌴ARES is an open-source framework for adaptive multimodal reasoning, using difficulty-aware training and entropy-shaped policy optimization. • 5 items • Updated Oct 12, 2025 • 2

published a dataset 4 months ago

ares0728/ARES-hard-validation

Viewer • Updated Oct 12, 2025 • 2.46k • 11

upvoted a collection 4 months ago

ARES

Collection

🌴ARES is an open-source framework for adaptive multimodal reasoning, using difficulty-aware training and entropy-shaped policy optimization. • 5 items • Updated Oct 12, 2025 • 2

updated a collection 4 months ago

ARES

Collection

🌴ARES is an open-source framework for adaptive multimodal reasoning, using difficulty-aware training and entropy-shaped policy optimization. • 5 items • Updated Oct 12, 2025 • 2

published a model 4 months ago

ares0728/ARES-Coldstart-7B

Updated Oct 12, 2025

updated a collection 4 months ago

ARES

Collection

🌴ARES is an open-source framework for adaptive multimodal reasoning, using difficulty-aware training and entropy-shaped policy optimization. • 5 items • Updated Oct 12, 2025 • 2

published a model 4 months ago

ares0728/ARES-RL-7B

Updated Oct 12, 2025 • 2