LCLMs are scaled 0.6B-encoder 4B-decoder compressors pre-trained on over 350B tokens that improve the Pareto frontier for general-task performance, compression speed, and peak memory in long-context language model inference.
Refrag: Rethinking rag based decoding
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 6verdicts
UNVERDICTED 6roles
background 2polarities
background 2representative citing papers
GroundedCache reduces unsafe-served rate in RAG answer caching to 0-1.5% (vs 15-51.5% naive) via four validation gates while keeping p50 latency within 1.07x of no-cache baseline.
Larger LLM compressors in lossy setups often yield less faithful context reconstructions due to knowledge overwriting and semantic drift, with mid-sized models outperforming larger ones across 27 tested configurations.
MiniPIC enables multiple position-independent caching methods inside vLLM via unrotated KV storage, per-request RoPE application, and three primitives, delivering 49% prefill throughput gains and up to 100x lower cached-span TTFT on 2WikiMultihopQA.
Merlin achieves byte-exact deduplication of text at up to 8.7 GB/s using SIMD-optimized hashing, reducing LLM context sizes by 13.9-71% with no data loss.
Byte-exact deduplication reduces RAG context size by 0.16% to 80.34% across three regimes with zero measurable quality regression per multi-vendor LLM evaluation.
citing papers explorer
-
End-to-End Context Compression at Scale
LCLMs are scaled 0.6B-encoder 4B-decoder compressors pre-trained on over 350B tokens that improve the Pareto frontier for general-task performance, compression speed, and peak memory in long-context language model inference.
-
Grounded Cache Routing for Retrieval-Augmented Generation: When Is It Safe to Reuse an Answer?
GroundedCache reduces unsafe-served rate in RAG answer caching to 0-1.5% (vs 15-51.5% naive) via four validation gates while keeping p50 latency within 1.07x of no-cache baseline.
-
When Less is More: The LLM Scaling Paradox in Context Compression
Larger LLM compressors in lossy setups often yield less faithful context reconstructions due to knowledge overwriting and semantic drift, with mid-sized models outperforming larger ones across 27 tested configurations.
-
MiniPIC: Flexible Position-Independent Caching in <100LOC
MiniPIC enables multiple position-independent caching methods inside vLLM via unrotated KV storage, per-request RoPE application, and three primitives, delivering 49% prefill throughput gains and up to 100x lower cached-span TTFT on 2WikiMultihopQA.
-
Merlin: Deterministic Byte-Exact Deduplication for Lossless Context Optimization in Large Language Model Inference
Merlin achieves byte-exact deduplication of text at up to 8.7 GB/s using SIMD-optimized hashing, reducing LLM context sizes by 13.9-71% with no data loss.
-
Byte-Exact Deduplication in Retrieval-Augmented Generation: A Three-Regime Empirical Analysis Across Public Benchmarks
Byte-exact deduplication reduces RAG context size by 0.16% to 80.34% across three regimes with zero measurable quality regression per multi-vendor LLM evaluation.