Geuntaek Lim

geuntaek's activity

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

upvoted 2 papers 4 days ago

Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs

Paper • 2610.36671 • Published 7 days ago • 124

Your Benchmark Is an Empirical Measure Over Difficulty

Paper • 2610.29927 • Published 13 days ago • 113

upvoted 2 papers 11 days ago

LaLa-World: Scaling Latent World Models with Natural Language Actions

Paper • 2610.27851 • Published 13 days ago • 51

Reading, Writing, Sharing, and Interacting with the World through Video

Paper • 2610.31668 • Published 12 days ago • 55

upvoted 2 papers 18 days ago

DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding

Paper • 2610.19849 • Published 20 days ago • 42

OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding

Paper • 2610.18829 • Published 21 days ago • 65

upvoted a paper 20 days ago

PhyProbe: Rethinking Physical Consistency Evaluation in Generated Videos

Paper • 2609.38377 • Published Sep 29 • 16

upvoted a paper 28 days ago

Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers

Paper • 2610.00531 • Published Sep 30 • 33

upvoted a paper 29 days ago

Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It

Paper • 2609.36585 • Published Sep 29 • 30

upvoted 4 papers about 1 month ago

OmniTaskonomy: When Does Visual Generation Improve Visual Understanding?

Paper • 2609.38079 • Published Sep 29 • 46

Knowing When Thinking Is Not Enough: Teaching Small Reasoning Models to Reason Beyond Their Parametric Knowledge

Paper • 2609.34327 • Published Sep 28 • 40

FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders

Paper • 2609.31620 • Published Sep 25 • 62

MLLMs Hallucinate when Information Distribution Drifts in Synergy Heads

Paper • 2609.09206 • Published Sep 5 • 11

upvoted 4 papers 3 months ago

ChronoVision: Temporal Reasoning via Latent State Reconstruction

Paper • 2608.05631 • Published Aug 6 • 18

Kimi K3: Open Frontier Intelligence

Paper • 2607.24653 • Published Jul 27 • 527

Self-Supervised Learning of Structured Dynamics from Videos

Paper • 2607.21576 • Published Jul 23 • 21

TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs

Paper • 2607.17423 • Published Jul 19 • 81

upvoted 3 papers 4 months ago

On-Policy Delta Distillation

Paper • 2607.15161 • Published Jul 16 • 36

4D Human-Scene Reconstruction from Low-Overlap Captures

Paper • 2607.09125 • Published Jul 10 • 51

Why Can't I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action Recognition

Paper • 2601.16211 • Published Jul 2 • 51