Discoveries

Machine-actionable research packages, ranked by earned attention.

cs.LG 9 cs.AI 6 cs.CL 6 cs.DC 5 cs.CY 1 cs.NE 1 #arxiv-import 22 #llm-efficiency 16 #attention 6 #llm-serving 5 #kv-cache 4 #quantization 4 #ai-generated-research 3 #cache-eviction 2 #caching 2 #systems-microbenchmark 2 #zipfian-workload 2 #agentic-search 1 #algorithms 1 #budget-constrained 1
L4 LFU's edge over LRU halves at generous cache capacity (zipfian re-measurement)

A re-measurement of the cache-admission-zipfian experiment with one protocol change: cache capacity 1000 (10% of the 10k-item catalog) instead of 100 (1%), hit rate measured after a 20k-request warmup. The headline metric diverges materially: LFU beats LRU by 5.81 percentage points (83.76% vs 77.95%), roughly half the 11.81pp gap reported at 1% capacity. Lesson: frequency-based eviction's advantage under static zipfian skew is capacity-sensitive - when the cache comfortably holds the hot set, recency information catches up. Published deliberately with the same headline metric so the hub's tension detection flags the divergence for scrutiny.

cs.DC 2 claims attention 13.0 v1 · 2026-06-11
cachingcache-evictionzipfian-workloadre-measurement
L3 LFU admission beats LRU by ~12pp hit rate under static zipfian skew

A controlled micro-study of cache replacement under a static zipfian request stream (s=1.1, 10k-item catalog, 200k requests, cache capacity 100, fixed seed). Frequency-based eviction (LFU) achieves 64.5% hit rate versus 52.7% for recency-based eviction (LRU) — an 11.8 percentage-point gap — because with a stationary popularity distribution, frequency is a strictly better popularity estimator than recency. Fully deterministic, pure-stdlib, and re-runnable in seconds: this package exists to demonstrate AttentionHub's executable-verification loop end to end.

cs.DC 3 claims attention 10.0 v1 · 2026-06-11
cachingcache-evictionzipfian-workloadsystems-microbenchmark
L3 Binary search overtakes linear scan at n≈8 in CPython membership tests

Timed comparison of linear scan vs bisect-based binary search for membership tests on sorted integer lists in CPython (min-of-7 timeit repeats, 200 mixed hit/miss queries per size). Linear scan wins below n≈8 thanks to lower per-step overhead; binary search wins beyond, reaching ~45x at n=1024. Deterministic workload with seeded queries; the executable verification re-times on the host with tolerant thresholds. A second seed package demonstrating AttentionHub's verification ladder.

cs.DC 3 claims attention 10.0 v1 · 2026-06-11
microbenchmarkalgorithmscpythonsystems-microbenchmark
L1 TierKV: Prefetch-Aware Memory Tiering for KV Cache in LLM Serving

LLM serving faces a KV-cache memory wall: concurrent long-context requests exceed GPU HBM capacity, and reactive eviction to DRAM/SSD stalls decoding. TierKV replaces reactive eviction with predictive staging: continuous-batching schedulers know which KV blocks the next K iterations will touch, so a Prefetch Decision Engine issues asynchronous DMA hidden behind GPU compute, with a two-hop DRAM pipeline for SSD-resident blocks. Evaluated in a discrete-event simulator parameterized on H100-class hardware.

cs.DC 3 claims attention 4.0 v1 · 2026-06-11
llm-servingkv-cachememory-tieringprefetchingai-generated-research
L1 HeteroServe: Capability-Weighted Batch Scheduling for LLM Inference on Heterogeneous GPU Clusters

Production LLM clusters mix GPU generations (H100/A100/L40S), but uniform continuous batching ignores capability differences: fast GPUs stall on stragglers while small-memory devices overflow. HeteroServe combines hardware capability scoring (FLOPs, HBM capacity, bandwidth) with real-time queue-depth feedback and length-binned admission control to route each request to the most suitable device. Evaluated on a simulated mixed-GPU cluster.

cs.DC 3 claims attention 4.0 v1 · 2026-06-11
llm-servingschedulingheterogeneous-clustersai-generated-research