会员服务
Witness Overlap: Directional Provenance Inside Open-Weight Model Families
Arxiv
0+阅读 · 10月15日
Tile-Based Phase Unwrapping with Reliability-Ordered Joining and Residual-Weighted Multipath Averaging
Arxiv
0+阅读 · 10月14日
Quality-Aware Self-Correcting Speech Translation on an Edge Device
Arxiv
0+阅读 · 10月14日
Natural Language Processing for Aligning Readings with Learning Objectives and Assessing Topic Balance in Dental Education
Arxiv
0+阅读 · 10月11日
Fast Trainable Multilinear Bases for Image Compression
Arxiv
0+阅读 · 10月11日
Arctan-skewing: a new skewing scheme on the hypertorus
Arxiv
0+阅读 · 10月9日
Two For The Price Of One: Curvature-Aware Pruning Preserves Influence Functions Too
Arxiv
0+阅读 · 10月13日
What Makes a Latent Space Shared? Direct and Recoverable Compatibility in Motion VAEs
Arxiv
0+阅读 · 10月12日
Where Decoder Cosine Similarity Fails for SAE Feature Flow Discovery
Arxiv
0+阅读 · 9月11日
WARP: Wasserstein-Aligned RAG for Population Opinions
Arxiv
0+阅读 · 9月22日
Gradients Know What Outcomes Don't: Unlocking Reinforcement Learning for LLM Reasoning with Gradient-Aligned Rewards
Arxiv
0+阅读 · 9月3日
BrachistoneLR: A Brachistochrone-Inspired Learning-Rate Schedule and a Controlled Benchmark of Scheduling Policies
Arxiv
0+阅读 · 9月8日
Nearest but Not Dearest: Shared Curator-Feedback Infrastructure for Content-Only Search and Recommendation
Arxiv
0+阅读 · 9月24日
ExpertMuon-Compass: Alignment-Guided Step Sizes for Mixture-of-Experts Training
Arxiv
0+阅读 · 10月2日
AdamX: Cosine similarity meets gradient descent
Arxiv
0+阅读 · 9月10日
参考链接
微信扫码咨询专知VIP会员
Top