A segment-wise soft compression method that provides up to 4x TTFT speedup, 2x KV cache reduction, and comparable or better QA quality at 2x compression.
Retaining key information under high compression ratios: Query-guided compressor for LLM s
1 Pith paper cite this work, alongside 3 external citations. Polarity classification is still indexing.
1
Pith paper citing it
3
external citations · OpenAlex
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CompLLM: Compression for Long Context Q&A
A segment-wise soft compression method that provides up to 4x TTFT speedup, 2x KV cache reduction, and comparable or better QA quality at 2x compression.