leo

duckking032's activity

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

upvoted a paper 3 days ago

Compile by Training: Turning Natural-Language Specifications into Local Neural Functions

Paper • 2609.04199 • Published 26 days ago • 332

upvoted 3 papers 4 days ago

Why Do Video Diffusion Models Violate Physics? Unveiling the Flaws in Attention Mechanisms

Paper • 2609.23658 • Published 9 days ago • 30

Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

Paper • 2609.29845 • Published 5 days ago • 86

Training Object Permanence in World Models

Paper • 2609.28654 • Published 6 days ago • 219

upvoted a paper 7 days ago

RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

Paper • 2609.24972 • Published 8 days ago • 212

upvoted a paper 11 days ago

DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression

Paper • 2609.19969 • Published 12 days ago • 189

upvoted a paper 19 days ago

Φ-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?

Paper • 2609.10226 • Published 20 days ago • 19

upvoted a paper 20 days ago

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

Paper • 2609.08183 • Published 21 days ago • 326

upvoted a paper 21 days ago

Unlocking Lossless Speedups in LLMs via Discrete Diffusion

Paper • 2609.04010 • Published 26 days ago • 114

upvoted 2 papers 22 days ago

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

Paper • 2609.05275 • Published 25 days ago • 26

MaxKernel: Agentic Kernel Generation for TPUs

Paper • 2609.04523 • Published 26 days ago • 20

upvoted a paper 25 days ago

Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments

Paper • 2609.04148 • Published 26 days ago • 247

upvoted a paper 26 days ago

SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers

Paper • 2609.01343 • Published 28 days ago • 91

upvoted a paper 28 days ago

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

Paper • 2608.30320 • Published 29 days ago • 63

upvoted a paper 30 days ago

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

Paper • 2607.05147 • Published Jul 6 • 51

upvoted 4 papers about 1 month ago

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

Paper • 2605.06885 • Published May 7 • 1

Prime Agent: A Self-Improving RLM Harness

Paper • 2608.23552 • Published Aug 24 • 56

ParaTempo: Efficient Parallel Reasoning via Temporal Confidence

Paper • 2608.16425 • Published Aug 17 • 41

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

Paper • 2305.07759 • Published May 12, 2023 • 49

upvoted a collection about 1 month ago

DFlash 2

Collection

Keep Drafting Parallel • 6 items • Updated Aug 28 • 48