CascadeDebate: Multi-Agent Deliberation for Cost-Aware LLM Cascades
On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks
Abstract (arXiv)
Cascaded LLM systems coordinate models of varying sizes with human experts to balance accuracy, cost, and abstention under uncertainty. However, single-model tiers at each stage often struggle with ambiguous queries, triggering premature escalations to costlier models or experts due to under-confidence and inefficient compute scaling. CascadeDebate addresses this gap by inserting multi-agent deliberation directly at each tier's escalation boundary. Confidence-based routers activate lightweight agent ensembles only for uncertain cases, enabling consensus-driven resolution of ambiguities internally without invoking higher-cost upgrades. Our unified architecture alternates single-model inference with selective multi-agent deliberation across model scales, culminating in human experts as the final fallback. This design scales test-time compute dynamically according to query difficulty. Across five benchmarks spanning science, medicine, and general knowledge, CascadeDebate outperforms strong single-model cascades and standalone multi-agent systems by up to 26.75 percent. An online threshold optimizer proves essential, boosting accuracy by 20.98 to 52.33 percent relative improvement over fixed policies and enabling elastic adaptation to real-world distributions.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 18 of its 26 references that are in the map. First posted on arXiv 14 Apr 2026.
Builds on
18 of 26 references.
Closest ancestors · 4
Same direction, newest first
- 2024A Unified Approach to Routing and Cascading for LLMs14 Oct
- 2024Mixture-of-Agents Enhances Large Language Model Capabilities7 Jun
- 2024Chain of Agents: Large Language Models Collaborating on Long-Context Tasks4 Jun
- 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar
Other references · 14
- 2020Measuring Massive Multitask Language Understandingcited by 7,768
- 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
- 2022Large Language Models Encode Clinical Knowledgecited by 1,807
- 2022Language Models (Mostly) Know What They Knowcited by 1,783
- 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
- 2020What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Examscited by 1,324
- 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 946
- 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 791
- 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
- 2024Hybrid LLM: Cost-Efficient and Quality-Aware Query Routingcited by 354
- 2023Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoningcited by 140
- 2025Multi-Agent Evolve: LLM Self-Improve through Co-evolutioncited by 58
Built on by
Citing papers per year: 2026: 3.
3 papers.
Similar, not cited
Close in wording, no citation either way
- 2026Sequential Consensus for Multi-Agent LLM Debates: A Wald-SPRT compute governor with calibration-based failure detection
- 2026Design and Evaluation of Multi-Agent AI Oracle Systems for Prediction Market Resolution
- 2026On the Reliability Limits of LLM-Based Multi-Agent Planning
- 2026TeamTR: Trust-Region Fine-Tuning for Multi-Agent LLM Coordination
- 2026COOPA: A Modular LLM Agent Architecture for Operations Research Problems
- 2026AgentHPOBench: A Benchmark For Evaluating LLM Agents as Sequential Hyperparameter Optimizers
- 2025HiveMind: Contribution-Guided Online Prompt Optimization of LLM Multi-Agent Systems
- 2025Debate Only When Necessary: Adaptive Multiagent Collaboration for Efficient LLM Reasoning