OmniEcho: Spatial Audio Understanding for Embodied Agents Paper • 2609.23407 • Published 9 days ago • 24
Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs Paper • 2609.29845 • Published 5 days ago • 80
Agora: Git as Shared Memory for Collective AutoResearch Paper • 2609.18094 • Published 13 days ago • 58
Confidence Comes from Experience: Experiential Confidence Estimation from Reasoning to Agents Paper • 2609.17708 • Published 14 days ago • 76
LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence Paper • 2609.17488 • Published 14 days ago • 809
Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction Paper • 2609.13285 • Published 21 days ago • 81
Dream-RSI: Recursive Self-Improvement through Evolving Worlds Paper • 2609.14858 • Published 15 days ago • 247
NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction Paper • 2609.10715 • Published 20 days ago • 329
WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data Paper • 2609.05405 • Published 25 days ago • 44
AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems Paper • 2609.08572 • Published 21 days ago • 107
Scaling Automatic Research Agents via World Models Paper • 2608.12564 • Published about 1 month ago • 482
Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation Paper • 2609.02998 • Published 27 days ago • 17
Unlocking Lossless Speedups in LLMs via Discrete Diffusion Paper • 2609.04010 • Published 26 days ago • 114
LatentPress: Context Compression Beyond Text and Vision Paper • 2609.01507 • Published 28 days ago • 121
Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning Paper • 2609.03430 • Published 26 days ago • 187
Terminal-Universe: Turning Agent Trajectories into Scalable Terminal Environments Paper • 2609.04148 • Published 26 days ago • 247
EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction Paper • 2609.02783 • Published 27 days ago • 62
Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills Paper • 2609.02749 • Published 27 days ago • 403
ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training Paper • 2609.00188 • Published 29 days ago • 54