Results for “file-compression”

Lossless LLM-optimized compression of source documents. Use when the user requests to 'distill documents' or 'create a distillate'.

Lossless LLM-optimized compression of source documents. Use when the user requests to 'distill documents' or 'create a distillate'.

Enables persistent ultra-compressed technical communication; use for explicit brevity or token-reduction requests, except where fragments risk safety or clarity.

Context compression for YAMTAM — nén JSON/structured tool output trước khi vào LLM. Hiệu quả với JSON (50-72% tiết kiệm); text thuần cần bản [all].

Compress large language models using knowledge distillation from teacher to student models, covering temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.