Training two LoRA adapters competitively against each other, where each reads the other's solution summary and is rewarded for out-solving it, doubles GRPO's pass@1 on hard math while shortening reasoning traces.
hub
Rethinking mixture-of-agents: Is mixing different large language models beneficial?
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
abstract
Ensembling outputs from diverse sources is a straightforward yet effective approach to boost performance. Mixture-of-Agents (MoA) is one such popular ensemble method that aggregates outputs from multiple different Large Language Models (LLMs). This paper raises the question in the context of language models: is mixing different LLMs truly beneficial? We propose Self-MoA -- an ensemble method that aggregates outputs from only the single top-performing LLM. Our extensive experiments reveal that, surprisingly, Self-MoA outperforms standard MoA that mixes different LLMs in a large number of scenarios: Self-MoA achieves $6.6\%$ improvement over MoA on the AlpacaEval 2.0 benchmark, and an average of $3.8\%$ improvement across various benchmarks, including MMLU, CRUX, and MATH. Applying Self-MoA to one of the top-ranking models in AlpacaEval 2.0 directly achieves the new state-of-the-art performance on the leaderboard. To understand the effectiveness of Self-MoA, we systematically investigate the trade-off between diversity and quality of outputs under various MoA settings. We confirm that the MoA performance is rather sensitive to the quality, and mixing different LLMs often lowers the average quality of the models. To complement the study, we identify the scenarios where mixing different LLMs could be helpful. This paper further introduces a sequential version of Self-MoA, that is capable of aggregating a large number of LLM outputs on-the-fly over multiple rounds, and is as effective as aggregating all outputs at once.
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Any single-output LLM ensemble is accuracy-capped at 1-beta where beta is the all-models-wrong rate, a quantity not captured by pairwise correlations and frequently underestimated by copula models.
A derived scaling law R(N) = 1/(1 + c(N-1)N^{-β}) fits answer diversity and correctness across 44 LLM multi-agent conditions with R² > 0.99, classifying regimes by β and showing only heterogeneous teams escape hard-ceiling saturation.
A bipartite factor graph with message-passing protocol and asymmetric damping aggregates multi-LLM predictions, cutting token use by 97% and API calls by 6X while outperforming baselines on MMLU, MMLU-Pro, GPQA, and MedMCQA.
Refute-or-Promote applies adversarial multi-agent review with kill gates and empirical verification to filter LLM defect candidates, killing 79-83% before disclosure and yielding 4 CVEs plus multiple accepted fixes across libraries, C++ standard, and compilers.
Pyramid MoA is a hierarchical Mixture-of-Agents system with a decision-theoretic router that achieves up to 42.9% compute savings while nearly matching oracle accuracy on MBPP, GSM8K, MMLU, HumanEval, and MATH.
SANet uses semantic-aware AI agents for cross-layer 6G optimization, achieving up to 14.61% performance gains with 44.37% of the FLOPs of prior methods via model partitioning and decentralized multi-objective algorithms.
ReM-MoA uses persistent ranked memory of reasoning traces and curated diversified routing to sustain depth scaling in Mixture-of-Agents, outperforming prior variants on five benchmarks with widening gains at greater depth.
SAGE compares social co-evolution against matched self-evolution across three arenas and finds peer history enables breakthroughs only for agents that plateau under self-improvement, with abstraction of traces mattering more than raw volume.
SIGMA builds a signed relational graph among LLM agents and uses conflict-aware message passing plus weighted aggregation to produce more consistent predictions than prior cooperative-assumption baselines.
LLM reliability techniques are unified as communication channel operators, with a new cost-aware router achieving superior quality-cost tradeoffs on hard tasks.
EP-HUBO treats CoT evidence selection as higher-order unconstrained binary optimization over per-hypothesis pools with quality weights to improve aggregation on legal benchmarks.
MOSAIC uses an Integer Linear Program scheduler for expert placement and prompt assignment plus adaptive aggregation to achieve 1.7-2.3x end-to-end speedup on 4-GPU MoA workloads while keeping accuracy within 0.1pp.
MAC framework selects Pareto-optimal LLM agents and masks low cross-consistency outputs for adaptive collaboration in medical decision-making.
SMCS coordinates 15 open-source LLMs via retrieval-based prior selection and exploration-exploitation posterior enhancement, outperforming GPT-4.1 by 5.36% and GPT-o3-mini by 5.28% on eight benchmarks.
Execution feedback in refinement loops improves 1-3B code generation performance far more than complex pipeline topologies discovered via evolutionary search on HumanEval and sanitized MBPP.
citing papers explorer
-
Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning
Training two LoRA adapters competitively against each other, where each reads the other's solution summary and is rewarded for out-solving it, doubles GRPO's pass@1 on hard math while shortening reasoning traces.
-
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
Any single-output LLM ensemble is accuracy-capped at 1-beta where beta is the all-models-wrong rate, a quantity not captured by pairwise correlations and frequently underestimated by copula models.
-
The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size
A derived scaling law R(N) = 1/(1 + c(N-1)N^{-β}) fits answer diversity and correctness across 44 LLM multi-agent conditions with R² > 0.99, classifying regimes by β and showing only heterogeneous teams escape hard-ceiling saturation.
-
From Talking Words to Sharing Thoughts: Scalable Multi-LLM Aggregation via Structured Message Passing
A bipartite factor graph with message-passing protocol and asymmetric damping aggregates multi-LLM predictions, cutting token use by 97% and API calls by 6X while outperforming baselines on MMLU, MMLU-Pro, GPQA, and MedMCQA.
-
Refute-or-Promote: An Adversarial Stage-Gated Multi-Agent Review Methodology for High-Precision LLM-Assisted Defect Discovery
Refute-or-Promote applies adversarial multi-agent review with kill gates and empirical verification to filter LLM defect candidates, killing 79-83% before disclosure and yielding 4 CVEs plus multiple accepted fixes across libraries, C++ standard, and compilers.
-
Pyramid MoA: A Probabilistic Framework for Cost-Optimized Anytime Inference
Pyramid MoA is a hierarchical Mixture-of-Agents system with a decision-theoretic router that achieves up to 42.9% compute savings while nearly matching oracle accuracy on MBPP, GSM8K, MMLU, HumanEval, and MATH.
-
SANet: A Semantic-aware Agentic AI Networking Framework for Cross-layer Optimization in 6G
SANet uses semantic-aware AI agents for cross-layer 6G optimization, achieving up to 14.61% performance gains with 44.37% of the FLOPs of prior methods via model partitioning and decentralized multi-objective algorithms.
-
ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling
ReM-MoA uses persistent ranked memory of reasoning traces and curated diversified routing to sustain depth scaling in Mixture-of-Agents, outperforming prior variants on five benchmarks with widening gains at greater depth.
-
SAGE: A Quantitative Evaluation of Socialized Evolution in Agent Ecosystems
SAGE compares social co-evolution against matched self-evolution across three arenas and finds peer history enables breakthroughs only for agents that plateau under self-improvement, with abstraction of traces mattering more than raw volume.
-
Conflict-Resilient Multi-Agent Reasoning via Signed Graph Modeling
SIGMA builds a signed relational graph among LLM agents and uses conflict-aware message passing plus weighted aggregation to produce more consistent predictions than prior cooperative-assumption baselines.
-
A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability
LLM reliability techniques are unified as communication channel operators, with a new cost-aware router achieving superior quality-cost tradeoffs on hard tasks.
-
Quantum-Inspired Trace-Augmented Evidence Selection for Reasoning over Structured Hypothesis Spaces
EP-HUBO treats CoT evidence selection as higher-order unconstrained binary optimization over per-hypothesis pools with quality weights to improve aggregation on legal benchmarks.
-
MOSAIC: Efficient Mixture-of-Agent Scheduling via Adaptive Aggregation and Inference Concurrency
MOSAIC uses an Integer Linear Program scheduler for expert placement and prompt assignment plus adaptive aggregation to achieve 1.7-2.3x end-to-end speedup on 4-GPU MoA workloads while keeping accuracy within 0.1pp.
-
MAC: Masked Agent Collaboration Boosts Large Language Model Medical Decision-Making
MAC framework selects Pareto-optimal LLM agents and masks low cross-consistency outputs for adaptive collaboration in medical decision-making.
-
A Scalable Multi-LLM Collaboration System with Retrieval-based Selection and Exploration-Exploitation-Driven Enhancement
SMCS coordinates 15 open-source LLMs via retrieval-based prior selection and exploration-exploitation posterior enhancement, outperforming GPT-4.1 by 5.36% and GPT-o3-mini by 5.28% on eight benchmarks.
-
Feedback Over Form: Why Execution Feedback Matters More Than Pipeline Topology in 1-3B Code Generation
Execution feedback in refinement loops improves 1-3B code generation performance far more than complex pipeline topologies discovered via evolutionary search on HumanEval and sanitized MBPP.
- The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes