When Words Predict Workload

• URL: http://arxiv.org/abs/2607.04951v1
• Date: Mon, 06 Jul 2026 11:27:29 GMT
• Title: When Words Predict Workload
• Abstract summary: We propose a CPU-side Linguistic Resource Forecasting (LRF) gateway to prevent hardware collapse.<n>The gateway extracts a 16-dimensional text-structure vector and applies an XGBoost predictor to forecast trap-band membership.<n>The predictor achieved a live-trial AUROC of $0.84$, and the dynamic $Tauroute(t)$ controller yielded an $8.2%$ relative reduction in misroutes compared to an equivalent static threshold.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Standard distributed \ac{llm} schedulers rely on static token counts or rolling latency averages, making them susceptible to failures on statutorily constrained text. On \ac{epo} claims governed by Article 84 \ac{epc}, linguistic rigidity makes human and machine authorship statistically indistinguishable. Resolving this ambiguity mid-flight forces dynamic multi-model ensemble expansion, triggering unpredictable KV-cache and weight-allocation spikes that saturate consumer-grade edge GPU VRAM and cause severe \ac{oom} crashes. To prevent hardware collapse, we propose a CPU-side Linguistic Resource Forecasting (LRF) gateway. The gateway extracts a 16-dimensional text-structure vector and applies an XGBoost predictor to forecast trap-band membership. The resulting escalation probability ($\Pesc$) is evaluated against a dynamic, closed-form routing threshold ($\Tauroute(t)$) computed via real-time latency telemetry. Requests are safely routed to either a local Qwen2.5-7B edge worker or a remote contrastive ensemble (Qwen2.5 7B + 32B) on an NVIDIA H100 \emph{before} any edge GPU memory is allocated. In a 6,000-request live trial, the LRF gateway reduced the operational misroute fraction ($R_{\mathrm{mis}}$) to $0.087$--$0.095$, an order of magnitude below the token-count baseline ($0.849$). Peak edge VRAM remained safely bounded at $\SI{4.82}{\gibi\byte}$ (under the $\SI{8}{\gibi\byte}$ ceiling) across a $27\times$ variation in \ac{wan} delay. The predictor achieved a live-trial AUROC of $0.84$, and the dynamic $\Tauroute(t)$ controller yielded an $8.2\%$ relative reduction in misroutes compared to an equivalent static threshold.

Related papers

• Drift-Aware LLM Routing with Sparse Contexts and Shared Budgets [0.0]
  A multi-model language service must route each request while preserving workload-level budgets for compute, latency, memory, or monetary cost.<n>We formulate nonstationary sparse contextual routing with multiple knapsack constraints and an optional shadow-audit stream.<n>The policy estimates reward and resource use from a rolling audit window, routes using pessimistic reward and optimistic cost estimates, updates resource shadow prices online, and applies a hard meter before commitment.
  arXiv  Detail & Related papers  (2026-09-01T03:39:59Z)
• Variance-Reduced Trajectory Unravelings for GPU Noisy Quantum-Circuit Simulation: Characterization and a Qiskit-Aer Integration Gap [0.0]
  We show that emphvariance-reduced unravelings -- projector and analog sampling -- sharply cut variance.<n>On a single consumer GPU, projector unraveling reaches a target standard error with $20.8times$ fewer trajectories than Qiskit-Aer's ttbatched_shots_gpu at $n=10$.<n>Because Aer's Born-rule collapse machinery already exists, we specify a minimal change that would unlock the technique in production.
  arXiv  Detail & Related papers  (2026-07-20T08:28:31Z)
• Design rules for fault-tolerant multi-gate teleportation [0.0]
  Multi-gate teleportation (MGT) packages $n$ remote gates into a single ebit via a 1-ebit fan-out quantum circuit.<n>We derive a design rule for fault-tolerant packet sizes, $nmaxtextcorr(d) = lceil d/2 rceil$ for rotated surface codes of distance.<n>We show that the standard MWPM decoder built from the packet circuit's noise model naturally corrects the correlated error.
  arXiv  Detail & Related papers  (2026-07-01T18:00:55Z)
• Scaling Neural Network Verification with Tensor Parallelism and Fully Sharded Data Parallelism [0.0]
  Formal neural network verification is bounded in practice by GPU memory.<n>We adapt two techniques originally developed for large-scale model training to the auto_LiRPA / $,$-CROWN verification framework.<n>Fully Sharded Data Parallelism (FSDP) shards only weight matrices with a per-layer AllGather, producing bounds that are bitwise identical to the single- GPU baseline.
  arXiv  Detail & Related papers  (2026-06-08T11:56:29Z)
• ChunkFT: Byte-Streamed Optimization for Memory-Efficient Full Fine-Tuning [58.54940026861599]
  textscChunkFT is a memory-efficient fine-tuning framework.<n>textscChunkFT enables gradient computation for arbitrary sub-tensors without modifying the network architecture.<n>textscChunkFT consistently outperforms existing memory-efficient baselines.
  arXiv  Detail & Related papers  (2026-05-20T13:44:44Z)
• Deterministic Realization of Classical Dissipation on Quantum Computers [0.0]
  Lattice Boltzmann (LB) on quantum devices must reconcile unitary gate evolution with the dissipative emph collide step.<n>In the multiple-relaxation-time (MRT) class, we work in the common setting of emphmodewise diagonal moment relaxation, $m_r'=_r,m_r$ with $_rin[-1,1]$.<n>The main result is the dissipative MRT block; construction of the equilibrium moment vector$mmathrmeq=Mfmathrm
  arXiv  Detail & Related papers  (2026-04-28T09:36:40Z)
• Complexity of Classical Acceleration for $\ell_1$-Regularized PageRank [14.919427330415608]
  We show that FISTA can improve the dependence on $$ while preserving the $1/$ locality scaling.<n>We analyze FISTA on a slightly over-regularized objective and show that, under a checkable confinement condition, all spurious activations remain inside a boundary set.<n>This yields a bound consisting of an accelerated $(sqrt)-1log(/varepsilon)$ term plus a boundary overhead $sqrtvol(mathcalB)/(3/2)$.
  arXiv  Detail & Related papers  (2026-02-24T17:35:46Z)
• Rank-Aware Spectral Bounds on Attention Logits for Stable Low-Precision Training [0.0]
  Attention scores in transformers are bilinear forms $S_ij = x_itop M x_j / sqrtd_h$ whose maximum magnitude governs overflow risk in low-precision training.<n>We derive a emphrank-aware concentration inequality: when the interaction matrix $M = WQ WKtop$ has rank $r ll d$, tail probabilities for $max_i,j|S_ij|$ decay as $exp(-d22/(
  arXiv  Detail & Related papers  (2026-02-21T14:29:22Z)
• A Reduction from Delayed to Immediate Feedback for Online Convex Optimization with Improved Guarantees [58.59385794080679]
  We introduce a continuous-time model under which regret decomposes into a delay-independent learning term and a delay-induced drift term.<n>For bandit convex optimization, we significantly improve existing regret bounds, with delay-dependent terms matching state-of-the-art first-order rates.
  arXiv  Detail & Related papers  (2026-02-02T18:17:34Z)
• Sharp Gap-Dependent Variance-Aware Regret Bounds for Tabular MDPs [54.28273395444243]
  We show that the Monotonic Value Omega (MVP) algorithm achieves a variance-aware gap-dependent regret bound of $$tildeOleft(left(sum_Delta_h(s,a)>0 fracH2 log K land mathttVar_maxtextc$.
  arXiv  Detail & Related papers  (2025-06-06T20:33:57Z)
• Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update [62.96781471194877]
  Two principled strategies for handling heavy-tailed noise, truncation and median-of-means, have been introduced to heavy-tailed bandits.<n>We propose a emphone-pass algorithm based on the online mirror descent framework.
  arXiv  Detail & Related papers  (2025-03-01T09:41:45Z)
• Sign Operator for Coping with Heavy-Tailed Noise in Non-Convex Optimization: High Probability Bounds Under $(L_0, L_1)$-Smoothness [74.18546828528298]
  We show that SignSGD with Majority Voting can robustly work on the whole range of complexity with $kappakappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappa
  arXiv  Detail & Related papers  (2025-02-11T19:54:11Z)
• On the $O(\frac{\sqrt{d}}{T^{1/4}})$ Convergence Rate of RMSProp and Its Momentum Extension Measured by $\ell_1$ Norm [54.28350823319057]
  This paper considers the RMSProp and its momentum extension and establishes the convergence rate of $frac1Tsum_k=1T.<n>Our convergence rate matches the lower bound with respect to all the coefficients except the dimension $d$.<n>Our convergence rate can be considered to be analogous to the $frac1Tsum_k=1T.
  arXiv  Detail & Related papers  (2024-02-01T07:21:32Z)
• Minimal Expected Regret in Linear Quadratic Control [79.81807680370677]
  We devise an online learning algorithm and provide guarantees on its expected regret. This regret at time $T$ is upper bounded (i) by $widetildeO((d_u+d_x)sqrtd_xT)$ when $A$ and $B$ are unknown.
  arXiv  Detail & Related papers  (2021-09-29T14:07:21Z)
• Asynchronous Stochastic Optimization Robust to Arbitrary Delays [54.61797739710608]
  We consider optimization with delayed gradients where, at each time stept$, the algorithm makes an update using a stale computation - d_t$ for arbitrary delay $d_t gradient. Our experiments demonstrate the efficacy and robustness of our algorithm in cases where the delay distribution is skewed or heavy-tailed.
  arXiv  Detail & Related papers  (2021-06-22T15:50:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.