C3PO: Optimized Large Language Model Cascades with Probabilistic Cost Constraints for Reasoning

• URL: http://arxiv.org/abs/2511.07396v1
• Date: Mon, 10 Nov 2025 18:50:27 GMT
• Title: C3PO: Optimized Large Language Model Cascades with Probabilistic Cost Constraints for Reasoning
• Authors: Antonios Valkanas, Soumyasundar Pal, Pavel Rumiantsev, Yingxue Zhang, Mark Coates,
• Abstract summary: Large language models (LLMs) have achieved impressive results on complex reasoning tasks, but their high inference cost remains a major barrier to real-world deployment.<n>Existing cascade methods rely on supervised training with labeled data, offer no theoretical generalization guarantees, and provide limited control over test-time computational cost.<n>We introduce C3PO, a self-supervised framework for optimizing LLM cascades under probabilistic cost constraints.
• Score: 24.65381108650337
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) have achieved impressive results on complex reasoning tasks, but their high inference cost remains a major barrier to real-world deployment. A promising solution is to use cascaded inference, where small, cheap models handle easy queries, and only the hardest examples are escalated to more powerful models. However, existing cascade methods typically rely on supervised training with labeled data, offer no theoretical generalization guarantees, and provide limited control over test-time computational cost. We introduce C3PO (Cost Controlled Cascaded Prediction Optimization), a self-supervised framework for optimizing LLM cascades under probabilistic cost constraints. By focusing on minimizing regret with respect to the most powerful model (MPM), C3PO avoids the need for labeled data by constructing a cascade using only unlabeled model outputs. It leverages conformal prediction to bound the probability that inference cost exceeds a user-specified budget. We provide theoretical guarantees on both cost control and generalization error, and show that our optimization procedure is effective even with small calibration sets. Empirically, C3PO achieves state-of-the-art performance across a diverse set of reasoning benchmarks including GSM8K, MATH-500, BigBench-Hard and AIME, outperforming strong LLM cascading baselines in both accuracy and cost-efficiency. Our results demonstrate that principled, label-free cascade optimization can enable scalable LLM deployment.

Related papers

• Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization [60.87651283510059]
  Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs.<n>We propose Dynamic Pruning Policy Optimization (DPPO), a framework that enables dynamic pruning while preserving unbiased gradient estimation.<n>To mitigate the data sparsity induced by pruning, we introduce Dense Prompt Packing, a window-based greedy strategy.
  arXiv  Detail & Related papers  (2026-03-04T14:48:53Z)
• Budget-Aware Anytime Reasoning with LLM-Synthesized Preference Data [57.996437077411315]
  We study the reasoning behavior of large language models (LLMs) under limited computation budgets.<n>We introduce an anytime reasoning framework and the Anytime Index, a metric that quantifies how effectively solution quality improves as reasoning tokens increase.<n> Experiments on NaturalPlan (Trip), AIME, and GPQA datasets show consistent gains across Grok-3, GPT-oss, GPT-4.1/4o, and LLaMA models.
  arXiv  Detail & Related papers  (2026-01-16T07:09:30Z)
• Conformal Constrained Policy Optimization for Cost-Effective LLM Agents [27.37909142846675]
  Large language models (LLMs) have recently made tremendous progress towards solving challenging AI problems.<n>We propose a novel strategy where we combine multiple LLM models with varying cost/accuracy tradeoffs in an agentic manner.<n>Our approach provides a principled and practical framework for deploying LLM agents that are significantly more cost-effective while maintaining reliability.
  arXiv  Detail & Related papers  (2025-11-14T19:39:28Z)
• Efficient LLM Collaboration via Planning [56.081879390960204]
  Small and large models take turns acting as planner and executor, exchanging plans in a multi-stage cascade to collaboratively solve tasks.<n>We demonstrate that COPE achieves performance comparable to large proprietary models, while drastically reducing the inference API cost.
  arXiv  Detail & Related papers  (2025-06-13T08:35:50Z)
• Plan and Budget: Effective and Efficient Test-Time Scaling on Large Language Model Reasoning [19.258292534503887]
  Plan-and-Budget is a model-agnostic, test-time framework that decomposes complex queries into sub-questions and allocates token budgets based on estimated complexity using adaptive scheduling.<n>Plan-and-Budget improves reasoning efficiency across a range of tasks and models, achieving up to +70% accuracy gains, tangential -39% token reduction, and +187.5% improvement in $E3$.
  arXiv  Detail & Related papers  (2025-05-22T01:56:29Z)
• Scalable Chain of Thoughts via Elastic Reasoning [61.75753924952059]
  Elastic Reasoning is a novel framework for scalable chain of thoughts.<n>It separates reasoning into two phases--thinking and solution--with independently allocated budgets.<n>Our approach produces more concise and efficient reasoning even in unconstrained settings.
  arXiv  Detail & Related papers  (2025-05-08T15:01:06Z)
• Supervised Optimism Correction: Be Confident When LLMs Are Sure [91.7459076316849]
  We establish a novel theoretical connection between supervised fine-tuning and offline reinforcement learning.<n>We show that the widely used beam search method suffers from unacceptable over-optimism.<n>We propose Supervised Optimism Correction, which introduces a simple yet effective auxiliary loss for token-level $Q$-value estimations.
  arXiv  Detail & Related papers  (2025-04-10T07:50:03Z)
• Translate Smart, not Hard: Cascaded Translation Systems with Quality-Aware Deferral [28.382040322550775]
  We propose a simple yet effective approach for machine translation using existing quality estimation (QE) metrics as deferral rules.<n>We show that QE-based deferral allows a cascaded system to match the performance of a larger model while invoking it for a small fraction.
  arXiv  Detail & Related papers  (2025-02-18T10:05:40Z)
• Reward-Guided Speculative Decoding for Efficient LLM Reasoning [80.55186052123196]
  We introduce Reward-Guided Speculative Decoding (RSD), a novel framework aimed at improving the efficiency of inference in large language models (LLMs)<n>RSD incorporates a controlled bias to prioritize high-reward outputs, in contrast to existing speculative decoding methods that enforce strict unbiasedness.<n>RSD delivers significant efficiency gains against decoding with the target model only, while achieving significant better accuracy than parallel decoding method on average.
  arXiv  Detail & Related papers  (2025-01-31T17:19:57Z)
• Agreement-Based Cascading for Efficient Inference [29.378710032811227]
  Agreement-Based Cascading (ABC) is a simple, effective adaptive inference technique.<n>ABC builds a cascade of models of increasing size/complexity, and uses agreement between ensembles of models at each level of the cascade as a basis for data-dependent routing.<n>We show that ABC can reliably act as a drop-in replacement for existing models and surpass the best single model it aims to replace in terms of both efficiency and accuracy.
  arXiv  Detail & Related papers  (2024-07-02T15:14:12Z)
• Improving Large Models with Small models: Lower Costs and Better Performance [81.55672406002715]
  We propose Data Shunt$+$ (DS$+$), a general paradigm for collaboration of small and large models. For instance, ChatGPT achieves an accuracy of $94.43%$ on Amazon Product sentiment analysis, and DS$+$ achieves an accuracy of $95.64%$, while the cost has been reduced to only $31.18%$.
  arXiv  Detail & Related papers  (2024-06-15T14:44:43Z)
• Model Cascading for Code: A Cascaded Black-Box Multi-Model Framework for Cost-Efficient Code Completion with Self-Testing [20.445496441396028]
  We introduce a novel framework combining model cascading and inference-time self-testing algorithms to find multiple near-optimal self-testing options on the cost-accuracy tradeoff.<n>Our approach leverages self-generated tests to both enhance accuracy and evaluate model cascading decisions.<n> Experimental results show that our cascading approach reduces costs by an average of 26%, and up to 70% in the best case.
  arXiv  Detail & Related papers  (2024-05-24T16:20:04Z)
• LLMC: Benchmarking Large Language Model Quantization with a Versatile Compression Toolkit [55.73370804397226]
  Quantization, a key compression technique, can effectively mitigate these demands by compressing and accelerating large language models. We present LLMC, a plug-and-play compression toolkit, to fairly and systematically explore the impact of quantization. Powered by this versatile toolkit, our benchmark covers three key aspects: calibration data, algorithms (three strategies), and data formats.
  arXiv  Detail & Related papers  (2024-05-09T11:49:05Z)

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.