EntroRouter: Learning Efficient Model Routing via Entropy Regulation

• URL: http://arxiv.org/abs/2606.29424v1
• Date: Sun, 28 Jun 2026 14:39:06 GMT
• Title: EntroRouter: Learning Efficient Model Routing via Entropy Regulation
• Abstract summary: We propose a single-round routing framework that treats entropy regulation as a core objective.<n>We demonstrate that Entro retains 98.3% of the strongest expert's accuracy while reducing computational costs by 48.25%.
• Score: 45.88530258803968
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Model routing balances solution accuracy and computational cost by selecting among models of varying capabilities. While recent multi-round frameworks interleave reasoning and planning, we identify a structural failure mode termed Trust Region Collapse. We demonstrate that the deep coupling of reasoning and routing, exacerbated by the dominance of strong pre-training priors under sparse supervision, leads to degenerate local optima where capable experts are systematically suppressed. To decouple these processes, we propose $\textbf{EntroRouter}$, a single-round routing framework that treats entropy regulation as a core objective. We first initialize the policy via Soft Supervision, fitting a distribution of suitable models to establish a high-entropy prior for exploration. Subsequently, we stabilize Reinforcement Learning using a Soft Anchor, which utilizes offline capability estimates to orchestrate controlled entropy contraction within a safe trust region. Extensive experiments demonstrate that EntroRouter retains 98.3% of the strongest expert's accuracy while reducing computational costs by 48.25%.

Related papers

• Adaptive Error Budget Allocation for Fault-Tolerant Quantum Resource Estimation: A Metaheuristic Approach [0.14323566945483496]
  System-level resource estimation is a key component of fault-tolerant quantum computing toolchains.<n>We propose a training-free optimization framework that performs derivative-free search directly on the Azure Quantum Resource Estor.<n>We evaluate our framework across 433 circuits spanning 2 to 91 qubits from 31 families in the MQT Bench.
  arXiv  Detail & Related papers  (2026-08-16T07:14:29Z)
• Meta-Attention: Bayesian Per-Token Routing for Efficient Transformer Inference [0.0]
  We propose Meta-Attention, a framework that dynamically routes each token to the most appropriate attention strategy.<n>Unlike prior routing approaches that use deterministic or prior-free learned routing, the Meta-Controller treats per-token mechanism selection as posterior inference.<n>This design produces principled routing uncertainty estimates that govern the soft-to-hard routing transition, mitigates routing collapse without ad hoc load-balancing losses, and yields better compute-performance trade-offs.
  arXiv  Detail & Related papers  (2026-05-27T12:21:28Z)
• Flexible Routing via Uncertainty Decomposition [13.854691350572821]
  Key strategy for balancing performance and cost in machine learning systems is to dynamically route queries to either a low-cost model or an expensive oracle.<n>We present a new uncertainty-aware router that avoids unnecessary oracle calls on inherently ambiguous queries.
  arXiv  Detail & Related papers  (2026-05-08T14:39:57Z)
• ODAR: Principled Adaptive Routing for LLM Reasoning via Active Inference [60.958331943869126]
  ODAR-Expert is an adaptive routing framework that optimize the accuracy-efficiency trade-off via principled resource allocation.<n>We show strong and consistent gains, including 98.2% accuracy on MATH and 54.8% on Humanity's Last Exam.
  arXiv  Detail & Related papers  (2026-02-27T05:22:01Z)
• Robust Optimization Approach and Learning Based Hide-and-Seek Game for Resilient Network Design [10.37026246853005]
  We study the design of resilient and reliable communication networks in which a signal can be transferred only up to a limited distance before its quality falls below an acceptable threshold.<n>When excessive signal degradation occurs, regeneration is required through regenerators installed at selected network nodes.
  arXiv  Detail & Related papers  (2026-02-12T11:48:24Z)
• Conformal Thinking: Risk Control for Reasoning on a Compute Budget [60.65072883773352]
  Reasoning Large Language Models (LLMs) enable test-time scaling, with dataset-level accuracy improving as the token budget increases.<n>We re-frame the budget setting problem as risk control, limiting the error rate while minimizing compute.<n>Our framework introduces an upper threshold that stops reasoning when the model is confident and a novel lower threshold that preemptively stops unsolvable instances.
  arXiv  Detail & Related papers  (2026-02-03T18:17:22Z)
• ResAD: Normalized Residual Trajectory Modeling for End-to-End Autonomous Driving [64.42138266293202]
  ResAD is a Normalized Residual Trajectory Modeling framework.<n>It reframes the learning task to predict the residual deviation from an inertial reference.<n>On the NAVSIM benchmark, ResAD achieves a state-of-the-art PDMS of 88.6 using a vanilla diffusion policy.
  arXiv  Detail & Related papers  (2025-10-09T17:59:36Z)
• Bayesian Mixture-of-Experts: Towards Making LLMs Know What They Don't Know [0.0]
  Mixture-of-Experts (MoE) has enabled the creation of massive yet efficient Large Language Models.<n>Standard deterministic routing mechanism presents a significant limitation.<n>This thesis proposes a structured textbfBayesian MoE routing framework.
  arXiv  Detail & Related papers  (2025-09-28T12:07:35Z)
• Duality and Policy Evaluation in Distributionally Robust Bayesian Diffusion Control [8.863520091178335]
  We consider a diffusion control problem of expected terminal numerical utility.<n>The controller imposes a prior distribution on the unknown drift of an underlying diffusion.<n>In practice, the prior will generally be incorrectly specified, and the degree of model misspecification can have a significant impact on policy performance.<n>We introduce a distributionally robust Bayesian control (DRBC) formulation in which the controller plays a game against an adversary who selects a prior in divergence neighborhood of a baseline prior.
  arXiv  Detail & Related papers  (2025-06-24T03:58:49Z)
• Centaur: Robust End-to-End Autonomous Driving with Test-Time Training [84.78837437133234]
  We propose Centaur, which updates a planner's behavior via test-time training without relying on hand-engineered rules or cost functions.<n>We develop a novel uncertainty measure, called Cluster Entropy, which is simple, interpretable, and compatible with state-of-the-art planning algorithms.
  arXiv  Detail & Related papers  (2025-03-14T17:59:41Z)
• PTEENet: Post-Trained Early-Exit Neural Networks Augmentation for Inference Cost Optimization [0.0]
  We describe a method for introducing "shortcuts" into the DNN feedforward inference process by skipping costly feedforward computations whenever possible.<n>The proposed method is based on the previously described BranchyNet (Teerapittayanon et al., 2016) and the EEnet (Demir,) architectures that jointly train the main network and early exit branches.<n>By defining adjusted thresholds on these confidence extensions, we can control in real-time the amount of data exiting from each branch and the overall tradeoff between speed and accuracy of our model.
  arXiv  Detail & Related papers  (2025-01-05T11:35:08Z)
• Do Wider Neural Networks Really Help Adversarial Robustness? [92.8311752980399]
  We show that the model robustness is closely related to the tradeoff between natural accuracy and perturbation stability. We propose a new Width Adjusted Regularization (WAR) method that adaptively enlarges $lambda$ on wide models.
  arXiv  Detail & Related papers  (2020-10-03T04:46:17Z)

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.