leo

duckking032's activity

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

upvoted a paper about 6 hours ago

Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference

Paper • 2609.05275 • Published 4 days ago • 12

upvoted a paper about 10 hours ago

MaxKernel: Agentic Kernel Generation for TPUs

Paper • 2609.04523 • Published 5 days ago • 8

upvoted a paper 3 days ago

Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments

Paper • 2609.04148 • Published 5 days ago • 276

upvoted a paper 4 days ago

SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers

Paper • 2609.01343 • Published 7 days ago • 98

upvoted a paper 6 days ago

On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability

Paper • 2608.30320 • Published 8 days ago • 53

upvoted a paper 8 days ago

DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation

Paper • 2607.05147 • Published Jul 6 • 49

upvoted a paper 11 days ago

Don't Retrain, Align: Adapting Autoregressive LMs to Diffusion LMs via Representation Alignment

Paper • 2605.06885 • Published May 7 • 1

upvoted 2 papers 12 days ago

Prime Agent: A Self-Improving RLM Harness

Paper • 2608.23552 • Published 15 days ago • 50

ParaTempo: Efficient Parallel Reasoning via Temporal Confidence

Paper • 2608.16425 • Published 21 days ago • 40

upvoted a paper 13 days ago

TinyStories: How Small Can Language Models Be and Still Speak Coherent English?

Paper • 2305.07759 • Published May 12, 2023 • 48

upvoted a collection 14 days ago

DFlash 2

Collection

Keep Drafting Parallel • 6 items • Updated 10 days ago • 43

upvoted 2 papers 14 days ago

Recursive Synthesis for Long-Horizon Terminal Tasks

Paper • 2608.05466 • Published Aug 7 • 252

Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts

Paper • 2608.20061 • Published 18 days ago • 46

upvoted a paper 18 days ago

On-Policy Self-Distillation without Any Supervision

Paper • 2608.06296 • Published 30 days ago • 218

upvoted a paper 19 days ago

FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution

Paper • 2608.16157 • Published 22 days ago • 107

upvoted a paper 25 days ago

BDH-CQ: In-Context Learning with Recurrent Latent Reasoning

Paper • 2608.09888 • Published 29 days ago • 775

upvoted 3 papers about 1 month ago

DAPD: Dual-Anchored Policy Distillation

Paper • 2608.01735 • Published Aug 3 • 153

AURORA-LM: Autoencoding Unified Representation for Continuous-Latent Diffusion Language Modeling

Paper • 2608.02602 • Published Aug 3 • 82

Kimi K3: Open Frontier Intelligence

Paper • 2607.24653 • Published Jul 27 • 515

upvoted a paper about 2 months ago

Loop the Loopies!

Paper • 2607.16051 • Published Jul 17 • 77