BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference
Large Reasoning Models (LRMs) achieve superior problem-solving through extended Chain-of-Thought (CoT) generation, but the resulting key-value (KV) cache grows linearly with sequence length and creates severe memory bottlenecks, often exceeding GPU capacity for long reasoning traces. Existing KV cache compression methods rely on recent queries to estimate future token importance, implicitly assuming these serve as reliable proxies for future attention patterns. We demonstrate that this assumption fails in long-horizon reasoning: certain decoding steps generate Thought Revisiting Tokens (TRT) t
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 57%Language Models Can Control Their Own Attention [R] →
- FuzzySimilar title/name (fuzzy) · 87%LMCache/LMCache →
“Fuzzy title match (0.94): “BeaconKV: Key-Value Cache Compression Guided by Beacon Queri” ≈ “LMCache/LMCache””
- FuzzySimilar title/name (fuzzy) · 84%xorbitsai/inference →
“Fuzzy title match (0.92): “BeaconKV: Key-Value Cache Compression Guided by Beacon Queri” ≈ “xorbitsai/inference””
- FuzzyOverlapping authors or contributors · 62%ultralytics/yolov5 →
“Shared author/contributor keys: choi”
- LinkedLinked via arxiv author · 85%Janghyeon Kim →
“BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference”
- LinkedLinked via arxiv author · 85%Minsoo Kim →
“BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference”
- LinkedLinked via arxiv author · 85%Kyuhong Shim →
“BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference”
- LinkedLinked via arxiv author · 85%Jungwook Choi →
“BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference”