会员服务
Learning is Forgetting: LLM Training As Lossy Compression
Arxiv
0+阅读 · 4月8日
ChipLight: Cross-Layer Optimization of Chiplet Design with Optical Interconnects for LLM Training
Arxiv
0+阅读 · 4月20日
HARP: Orchestrating Automated Parallel Training on Heterogeneous GPU Clusters
Arxiv
0+阅读 · 5月4日
FlashOverlap: Minimizing Tail Latency in Communication Overlap for Distributed LLM Training
Arxiv
0+阅读 · 4月27日
OT on the Map: Quantifying Domain Shifts in Geographic Space
Arxiv
0+阅读 · 4月17日
WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling
Arxiv
0+阅读 · 4月22日
AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism
Arxiv
0+阅读 · 4月29日
MatchRDMA: A Segmented and Rate-Matched Long-Haul RDMA Scheme for Geo-distributed LLM Training over OTN
Arxiv
0+阅读 · 4月27日
FedPLT: Scalable, Resource-Efficient, and Heterogeneity-Aware Federated Learning via Partial Layer Training
Arxiv
0+阅读 · 5月4日
Pause or Fabricate? Training Language Models for Grounded Reasoning
Arxiv
0+阅读 · 4月21日
MegaScale-Data: Scaling Dataloader for Multisource Large Foundation Model Training
Arxiv
0+阅读 · 4月27日
AHCQ-SAM: Toward Accurate and Hardware-Compatible Post-Training Segment Anything Model Quantization
Arxiv
0+阅读 · 4月8日
Curiosity-Critic: Cumulative Prediction Error Improvement as a Tractable Intrinsic Reward for World Model Training
Arxiv
0+阅读 · 4月29日
ProTrain: Efficient LLM Training via Memory-Aware Techniques
Arxiv
0+阅读 · 4月20日
TACO: Efficient Communication Compression of Intermediate Tensors for Scalable Tensor-Parallel LLM Training
Arxiv
0+阅读 · 4月27日
参考链接
微信扫码咨询专知VIP会员
Top