New techniques such as FPO, BCMT, ECASQ, and depth-aware MoE masking aim to reduce memory, compute, and training overhead while preserving model quality. These methods support more practical local adaptation and on-premise inference.
ECASQ jointly optimizes adaptive quantization and lossless compression by minimizing MSE under an entropy budget and an unbiasedness constraint. The optimal dynamic program runs in O(sd^2) time and O(d^2) space. A GPU-friendly approximate version red...
FPO proposes fine-tuning LLMs without propagating errors between layers or building autograd graphs. The method reduces peak training memory and increases throughput, but concentrates adaptation in the final layers. For self-hosted deployments, the o...
FPO adapts LLMs without a backward pass through the model body, reaching 2.7–3.2x the throughput of standard fine-tuning and about 40% less peak training memory. On OLMo-2-7B, Qwen3-8B, and Falcon3-7B, it improves in-domain perplexity while leaving M...
BCMT separates local token interaction from global context propagation. In tests up to 1024 tokens, it achieves validation performance comparable to Dense Transformers, with higher training throughput and lower memory consumption. The exponential cau...
A study on Qwen3.6-35B-A3B shows that sensitivity to expert masking in MoE models depends heavily on depth. Early and middle layers are fragile, while late layers tolerate aggressive cuts. On H100 servers, targeted late-layer policies preserve far mo...
