Junseok Kim

junseokkim00's activity

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

upvoted 2 papers 5 months ago

Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training

Paper • 2509.25758 • Published Sep 30, 2025 • 25

The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models

Paper • 2605.06196 • Published May 7 • 9

upvoted an article 5 months ago

view article

Article

Navigating the RLHF Landscape: From Policy Gradients to PPO, GAE, and DPO for LLM Alignment

• 

NormalUhr

•

Feb 11, 2025

• 135

upvoted 2 papers 5 months ago

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs

Paper • 2601.11061 • Published Jan 16 • 8

Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting

Paper • 2601.02151 • Published Jan 5 • 92

upvoted a paper 9 months ago

CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs

Paper • 2406.18521 • Published Jun 26, 2024 • 31

upvoted an article 11 months ago

view article

Article

Activation Steering: A New Frontier in AI Control—But Does It Scale?

• 

royswastik

•

Feb 2, 2025

• 6

upvoted a paper about 1 year ago

Deep Think with Confidence

Paper • 2508.15260 • Published Aug 21, 2025 • 92

upvoted an article over 2 years ago

view article

Article

Mixture of Experts Explained

• 
• 
• +4

osanseviero, lewtun, philschmid, smangrul, ybelkada, pcuenq

•

Dec 11, 2023

• 1.19k

upvoted a collection over 2 years ago

Agent

Collection

113 items • Updated Jan 15 • 22