The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary

• URL: http://arxiv.org/abs/2606.00376v1
• Date: Fri, 29 May 2026 21:35:23 GMT
• Title: The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary
• Abstract summary: Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks.<n>We show that tool-integrated reasoning consistently outperforms neural chain-of-thought.<n>Our results provide principled guidance for when pure neural reasoning should yield to hybrid approaches in agentic systems.
• Score: 13.891522069967507
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Extended chain-of-thought reasoning can degrade performance on deterministic state-tracking tasks, not due to preference biases, but limits rooted in the information-theoretic capacity of decoder-only attention. We establish: (1) an Attention Bottleneck Theorem with a complementary achievability construction, bounding state-tracking capacity as $O(H \cdot \log(L/H) \cdot \sqrt{d_h})$; (2) a context-dependent error model yielding super-exponential accuracy decay; (3) the State-Space Jaccard metric distinguishing capability from preference failures; (4) a Deterministic Horizon $d^* \in [19, 31]$ beyond which tool delegation becomes necessary. Across 12 models and 8 task domains (including SWE-Bench, WebArena, and SQL-Multi), tool-integrated reasoning consistently outperforms neural chain-of-thought; on the primary model suite it reaches 86-94% accuracy versus 24-42% for neural chain-of-thought. Fine-tuning on optimal-length traces yields $<$5% improvement, confirming an architectural ceiling, and high cross-model correlation ($r = 0.81$-$0.91$) indicates these failures are architectural rather than training-specific. Our results provide principled guidance for when pure neural reasoning should yield to hybrid approaches in agentic systems.

Related papers

• Agentic Stage-One Stellarator Optimization: Autonomous Multi-Objective Search for Finite-Beta Equilibria [0.0]
  We present a proof of concept for emphagentic stage-one optimization.<n>Agentic outer-loop control can sustain finite-beta, multi-objective search and turn repeated optimization into a scalable source of improved equilibria.
  arXiv  Detail & Related papers  (2026-08-02T16:08:52Z)
• Quantum Circuit Vision: Cost-Aware Evaluation of Visual AI Agents for Quantum Code Generation [0.8652572023509929]
  We present Quantum Circuit Vision, a cost-aware evaluation framework for multimodal AI agents on quantum circuit visual understanding.<n>We construct a 132-circuit benchmark spanning 13 categories ($1$--$10$ qubits) with executable Amazon Braket code and unitary-fidelity verification.<n>We propose a cascade routing strategy that achieves 84% accuracy at 38% of single-model cost, demonstrating that model routing dominates prompt engineering as a cost lever.
  arXiv  Detail & Related papers  (2026-07-11T00:54:52Z)
• Relative Kinetic Utility for Reasoning-Aware Structural Pruning in Large Language Models [0.0]
  We propose a novel framework that elevates discrete pruning to a continuous kinetic integral over the depth manifold of a model.<n>Experiments on Qwen-2.5-7B and LLaMA-3-8B improve performance in the high-sparsity regime around 40%.<n>RKU attains 13.34% accuracy on GSM8K at 40% sparsity, outperforming the strongest baseline.
  arXiv  Detail & Related papers  (2026-05-09T15:47:53Z)
• Teaching Thinking Models to Reason with Tools: A Full-Pipeline Recipe for Tool-Integrated Reasoning [59.74608632210439]
  We show how to inject natural tool-use behavior into a strong thinking model without sacrificing its no-tool reasoning ability.<n>Our recipe yields models that achieve state-of-the-art performance in a wide range of benchmarks among open-source models.
  arXiv  Detail & Related papers  (2026-05-07T14:23:21Z)
• Evaluating Small Language Models for Front-Door Routing: A Harmonized Benchmark and Synthetic-Traffic Experiment [0.05586191108738562]
  Small language models (SLMs) have sufficient reasoning capability for sub-second, zero-marginal-cost, self-hosted task classification.<n>Study 1 is a harmonized offline benchmark of Phi-3.5-mini, Qwen2.5-1.5B, and Qwen-2.5-3B on identical Azure T4 hardware, serving stack, quantization, and a fixed 60-case corpus.<n>Study 2 is a pre-registered four-arm randomized experiment under synthetic traffic with an effective sample size of 60 unique cases per arm.
  arXiv  Detail & Related papers  (2026-03-26T15:57:46Z)
• Reinforcement Learning from Multi-Source Imperfect Preferences: Best-of-Both-Regimes Regret [71.69884486156359]
  We study episodic RL from emphmulti-source imperfect preferences through a cumulative imperfection budget.<n>We propose a unified algorithm with regret $tildeO(sqrtK/M+)$, which exhibits a best-of-both-regimes behavior.
  arXiv  Detail & Related papers  (2026-03-20T19:34:53Z)
• Differentiable Symbolic Planning: A Neural Architecture for Constraint Reasoning with Learned Feasibility [0.0]
  Differentiable Symbolic Planning (DSP) is a neural architecture that performs discrete symbolic reasoning.<n>We integrate DSP into a Universal Cognitive Kernel (UCK) that combines graph attention with iterative constraint propagation.<n>UCK+DSP achieves 97.4% accuracy on planning under 4x size generalization.
  arXiv  Detail & Related papers  (2026-02-19T03:38:03Z)
• ReLE: A Scalable System and Structured Benchmark for Diagnosing Capability Anisotropy in Chinese LLMs [37.23311145049677]
  We present ReLE, a scalable system designed to diagnose Capability Anisotropy.<n>We evaluate 304 models across a Domain $times$ Capability Symbolic matrix comprising 207,843 samples.
  arXiv  Detail & Related papers  (2026-01-24T09:57:59Z)
• Towards a Science of Scaling Agent Systems [79.64446272302287]
  We formalize a definition for agent evaluation and characterize scaling laws as the interplay between agent quantity, coordination structure, modelic, and task properties.<n>We derive a predictive model using coordination metrics, that cross-validated R2=0, enabling prediction on unseen task domains.<n>We identify three effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead, and (2) a capability saturation: coordination yields diminishing or negative returns once single-agent baselines exceed 45%.
  arXiv  Detail & Related papers  (2025-12-09T06:52:21Z)
• vAttention: Verified Sparse Attention [100.98210818821688]
  vAttention is a practical sparse attention mechanism with user-specified $(epsilon, delta)$ guarantees on approximation accuracy (thus, verified)<n>We show that vAttention significantly improves the quality of sparse attention across datasets.<n>It can be deployed in reasoning scenarios to achieve fast decoding without compromising model quality.
  arXiv  Detail & Related papers  (2025-10-07T08:46:08Z)
• Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning [53.45095336430027]
  We develop a unified framework that combines implicit retrieval and structured collaboration.<n>On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3% accuracy.<n>Results on SuperGPQA and TRQA confirm robustness across domains.
  arXiv  Detail & Related papers  (2025-09-25T14:05:55Z)
• Understanding, Predicting and Better Resolving Q-Value Divergence in Offline-RL [86.0987896274354]
  We first identify a fundamental pattern, self-excitation, as the primary cause of Q-value estimation divergence in offline RL. We then propose a novel Self-Excite Eigenvalue Measure (SEEM) metric to measure the evolving property of Q-network at training. For the first time, our theory can reliably decide whether the training will diverge at an early stage.
  arXiv  Detail & Related papers  (2023-10-06T17:57:44Z)
• Self-Evaluation Guided Beam Search for Reasoning [61.523627290397556]
  We introduce a stepwise self-evaluation mechanism to guide and calibrate the reasoning process of Large Language Model (LLM) We propose a decoding algorithm integrating the self-evaluation guidance via beam search. Our approach surpasses the corresponding Codex-backboned baselines in few-shot accuracy by $6.34%$, $9.56%$, and $5.46%$ on the GSM8K, AQuA, and StrategyQA.
  arXiv  Detail & Related papers  (2023-05-01T02:37:59Z)
• NPBDREG: A Non-parametric Bayesian Deep-Learning Based Approach for Diffeomorphic Brain MRI Registration [0.0]
  NPBDREG is a non-optimal framework for unsupervised deformable image registration. It provides improved uncertainty estimates and confidence measures in a theoretically well-parametric and computationally efficient way. It shows a slight improvement in the registration accuracy compared to PrVXM.
  arXiv  Detail & Related papers  (2021-08-15T16:00:06Z)

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.