DLoop: Looped Speculative Decoding
Paper • 2610.07659 • Published • 11
SeongWan Kim
idgmatrix
AI & ML interests
None yet
Recent Activity
upvoted a paper about 10 hours ago
DLoop: Looped Speculative Decoding upvoted a paper about 10 hours ago
Improving Proactive AI Assistance with Hierarchical Procedural Understanding upvoted a paper about 10 hours ago
On-Policy Distillation with Negative-Policy RolloutsOrganizations
None yet
upvoted 6
papers about 10 hours ago
Improving Proactive AI Assistance with Hierarchical Procedural Understanding
Paper • 2610.06505 • Published • 9
On-Policy Distillation with Negative-Policy Rollouts
Paper • 2610.07874 • Published • 16
Tetris3D: 3D Scene Generation With Objects That Fit Together
Paper • 2610.10539 • Published • 35
RunningTab: Direct Workspace Interaction with Environment-Side Tabs
Paper • 2610.10444 • Published • 34
GRACE: Generation-aware latent compression for efficient video generation
Paper • 2610.10524 • Published • 60
upvoted 5
papers 2 days ago
Code2Games: Enabling Coding Agents for Gaming World Generation
Paper • 2610.05033 • Published • 7
In-Distribution Forcing for Long Video Generation at Test Time
Paper • 2610.03120 • Published • 43
RobotUse: Allocating Computation, Context, and Decisions
Paper • 2610.04929 • Published • 19
When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models
Paper • 2610.05719 • Published • 24
CANOPY: Adaptive-Granularity Evidence Compression for Multimodal RAG
Paper • 2610.00923 • Published • 21
upvoted 9
papers 3 days ago
The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks
Paper • 2609.25804 • Published • 164
On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics
Paper • 2609.35259 • Published • 198
Source Preference in the Wild: How LLM Agents Favor Items by Source, and How to Reduce It
Paper • 2610.03195 • Published • 44
Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers
Paper • 2610.00531 • Published • 58
Learning What to Recall: Adaptive Multi-Cue Episodic Memory for World Models
Paper • 2609.34677 • Published • 12
Beyond the Current Scene: Event-Referential Grasping with Active View Selection
Paper • 2609.39375 • Published • 51
World Observer: Joint Actor-Observer Generation for Persistent World Modeling
Paper • 2610.02162 • Published • 87
Overcoming Scaling Limits in On-Policy Self-Distillation for LLM Reasoning
Paper • 2609.37915 • Published • 8
A2Z GameSpec-Bench: How Faithfully Can Coding Agents Generate Games from Game Design Specifications?
Paper • 2609.39564 • Published • 17