RI

#1MuEvo: LLM-Driven Evolution of Multi-Heuristic Ensemble

2026-08-04 | Southern University of Science and Technology, The Hong Kong University of Science and Technology |2608.03636

MuEvo introduces an LLM-driven co-evolutionary framework that simultaneously optimizes multiple interacting heuristics within an algorithmic solver (like ACO or Selection Hyper-Heuristics) rather than isolating a single component. The results are backed by extensive numerical evidence across TSP, CVRP, BPP and Flowshop, showing consistent improvements over human-designed defaults and single-component LLM-AHD baselines (e.g., reducing the TSP-L optimality gap from 6.26% to 5.24%). KEY INSIGHT: To prevent a component from overfitting to a specific set of collaborators, candidate heuristics should be evaluated across multiple contexts (e.g., the current best ensemble, the initial ensemble and a diverse secondary ensemble), while an LLM summarizes cross-component interaction history to guide joint pair evolution.

#1ElastiCo: Elastic Configuration and Interference-Aware Orchestration for GPU Clusters

2026-08-08 | Beihang University, University of Leeds, University of Sydney |2608.07971

ElastiCo is a GPU cluster scheduler that co-locates deep learning training and offline LLM inference by treating job configurations as elastic variables and using shadow pricing for multi-resource allocation. The results are backed by strong empirical evidence on a 64-GPU testbed and 512-GPU simulation, demonstrating a 2.94x reduction in average job completion time and an increase in GPU utilization from 25% to 46% compared to static partitioning baselines. The key insight is the combination of 'Resource Shape Transformation'—exposing a family of valid resource-performance profiles for each job—with a Lagrangian decomposition that uses dynamic shadow prices to resolve cluster-wide multi-resource contention.

#2Hybrid-Adaptive Thread Tuning to Mitigate Simulation Execution Bottlenecks in High-Performance Reinforcement Learning Inference

2026-08-06 | National University of Defense Technology |2608.06025

Su et al. propose AutoThread, a hybrid adaptive thread-tuning method that dynamically optimizes worker thread counts for reinforcement learning simulation environments. The authors demonstrate real empirical gains, achieving up to an 83.8% runtime reduction and 1.8x throughput improvement over baseline tuning methods on AMD and Intel platforms. The key insight is the use of a Physics-Informed Neural Operator (PINO) where a finite-source M/M/1 queueing model acts as a structural constraint on the neural network's loss function, ensuring predictions remain physically plausible under dynamic workloads.

#3When Does Disaggregation Pay? Simulating Prefill--Decode--Attention--FFN Specialization for Agentic LLM Inference

2026-08-04 | Imperial College London, University of Cambridge |2608.03741

Forys et al. introduce HeteroPanacea, a simulation framework to evaluate 4-way (Prefill-Decode-Attention-FFN) disaggregated LLM serving across heterogeneous hardware and parallelization strategies. Backed by extensive simulation data, they demonstrate that this PDAF disaggregation yields up to a 2.06x throughput gain over non-disaggregated serving for prefill-heavy agentic workloads on custom NPUs. The key insight is that the hardware demands of attention (KV-bandwidth bound) and FFN (compute/weight bound) diverge so sharply in agentic workloads that they require entirely different hardware profiles to avoid resource stranding.

#4Every Cache Entry Earns Its Place: Global Allocation of Resolution and Coverage for KV Cache Compression

2026-08-07 | Tsinghua University |2608.07001

GraceKV formulates KV cache compression for long-context LLMs as a global resource allocation problem, using prototype trees and greedy marginal-utility allocation to balance local resolution and broad coverage. The method is backed by strong empirical results, achieving state-of-the-art performance across 24 of 32 settings on LongBench and RULER while reducing KV cache memory by up to 92% at 30K context lengths. The key insight is treating KV cache compression not as a fixed eviction or merging rule, but as a unified knapsack-like allocation problem across layers, heads and context slots, enabling dynamic, input-conditioned memory optimization.

#5BALANCE: Hybrid Autoregressive-Speculative LLM Inference in Wireless Edge Networks

2026-08-06 | The University of Hong Kong, Imperial College London |2608.05926

Qu et al. propose an edge LLM inference scheduling framework that maximizes task throughput by jointly assigning users to either autoregressive or speculative decoding while partitioning GPU memory and compute. The results are backed by numerical simulations and device measurements, demonstrating ~30-38% throughput improvements over single-mode baselines. The key insight is that the latency-memory trade-off between AD (low memory, high latency) and SD (high memory, low latency) can be effectively optimized by decoupling the modes and applying greedy knapsack heuristics based on user-specific latency contributions.