DICE: Entropy-Regularized Equilibrium Selection for Stable Multi-Agent LLM Coordination

• URL: http://arxiv.org/abs/2606.08068v1
• Date: Sat, 06 Jun 2026 09:33:32 GMT
• Title: DICE: Entropy-Regularized Equilibrium Selection for Stable Multi-Agent LLM Coordination
• Abstract summary: Multi-agent large language model (LLM) systems often fail to reliably outperform a single strong model equipped with best-of-N sampling.<n>We formalize a broad class of such systems as discounted incomplete-information Markov games.<n>We show that two common pathologies, oscillation between competing conventions and drift across them, can both induce unstable learning and linear Bayesian regret.
• Score: 33.199914481547346
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multi-agent large language model (LLM) systems often fail to reliably outperform a single strong model equipped with best-of-N sampling. We argue that a core source of this instability is ill-posed equilibrium selection: current systems specify what information agents share, but not which coordination convention should be selected. We formalize a broad class of such systems as discounted incomplete-information Markov games and show that two common pathologies, oscillation between competing conventions and drift across them, can both induce unstable learning and linear Bayesian regret. To obtain a well-posed target, we introduce the Heterogeneous Quantal Response Equilibrium (HQRE), an entropy-regularized equilibrium concept with agent- and state-dependent temperatures. Under a monotonicity condition, HQRE is unique, admits linearly convergent mirror updates, and yields bounded Bayesian regret; the same condition yields rollout-measurable stability diagnostics. We instantiate this objective in two algorithms: DICE-PC, which coordinates frozen models through prompt-control actions, and DICE-FT, which performs parameter-efficient mirror fine-tuning. Across eleven benchmarks in four domains, DICE improves accuracy-cost trade-offs over strong within-class baselines; on reasoning and planning tasks, DICE-PC improves by 4.3 percentage points on average and DICE-FT by 8.5 points.

Related papers

• Ensemble Diversity Optimization for Subjective Supervision [4.032367157209129]
  We introduce Ensemble Diversity Optimization (EDO), a prediction-space framework that jointly optimize ensemble weights, effective cardinality, and calibration.<n>EDO learns ensemble composition and size end-to-end via Gumbel-Softmax relaxation and incorporates a signed diversity regularizer.
  arXiv  Detail & Related papers  (2026-07-09T13:53:02Z)
• Efficient Conditioning Why Pseudo Observation Batch Bayesian Optimization Works When It Does not [0.0]
  Constant Liar (CL), Kriging Believer (KB), and fantasy models are widely used for batch selection in parallel Bayesian Optimization.<n>We identify efficient conditioning as the key surrogate property the ability to update predictions in closed form when data is augmented.<n>We unify CL, KB, and fantasy models as instances of a single conditioning mechanism differing only in lie value distribution.
  arXiv  Detail & Related papers  (2026-05-12T13:14:24Z)
• Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibration [72.0672328514289]
  Multimodal learning often grapples with the challenge of low-quality data.<n>We propose a unified framework, termed Conformal Predictive Self-Calibration.<n>Our framework consistently outperforms existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-05T14:48:52Z)
• Learning from Emptiness: De-biasing Listwise Rerankers with Content-Agnostic Probability Calibration [76.08899010904652]
  CapCal is a training-free framework that mechanically decouples positional bias from ranking decisions.<n>It achieves superior performance among training-free methods while preserving single-pass efficiency.
  arXiv  Detail & Related papers  (2026-04-11T10:47:22Z)
• Diffusion Controller: Framework, Algorithms and Parameterization [54.82539154511621]
  We present a unified control-theoretic view that casts reverse diffusion sampling as state-only control within (generalized) linearly-solvable Markov Decision Processes.<n>Under this framework, control acts by reweighting the pretrained reverse-time transition kernels, balancing terminal objectives against an $f$-divergence cost.<n>Experiments on Stable Diffusion v1.4 show consistent gains in preference-alignment win rates and improved quality-efficiency trade-offs.
  arXiv  Detail & Related papers  (2026-03-07T01:49:59Z)
• Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs [55.77845440440496]
  Push-based decentralized communication enables optimization over communication networks, where information exchange may be asymmetric.<n>We develop a unified uniform-stability framework for the Gradient Push (SGP) algorithm.<n>A key technical ingredient is an imbalance-aware generalization bound through two quantities.
  arXiv  Detail & Related papers  (2026-02-24T05:32:03Z)
• Mean-Field Path-Integral Diffusion: From Samples to Interacting Agents [0.0]
  Mean-Field Path-Integral Diffusion (MF-PID) is a framework in which samples are promoted to interacting agents whose drift depends self-consistently on the evolving population density.<n>MF-PID achieves 19--24% reductions in cumulative control energy over independent-agent baselines while matching the prescribed terminal distribution exactly.
  arXiv  Detail & Related papers  (2026-02-23T21:14:06Z)
• Consistency Deep Equilibrium Models [8.278751626877431]
  Deep Equilibrium Models (DEQs) have emerged as a powerful paradigm in deep learning.<n>DEQs incur significant inference latency due to the iterative nature of fixed-point solvers.<n>We introduce the Consistency Deep Equilibrium Model (C-DEQ) to accelerate DEQ inference.
  arXiv  Detail & Related papers  (2026-02-03T02:42:48Z)
• Deep Time-series Forecasting Needs Kernelized Moment Balancing [56.619037429652984]
  Deep time-series forecasting can be formulated as a distribution balancing problem aimed at aligning the distribution of the forecasts and ground truths.<n>We propose direct forecasting with kernelized moment balancing (KMB-DF)<n>Experiments across multiple models and datasets show that KMB-DF consistently improves forecasting accuracy and achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-01-31T13:20:18Z)
• EmoLoom-2B: Fast Base-Model Screening for Emotion Classification and VAD with Lexicon-Weak Supervision and KV-Off Evaluation [14.702916401734916]
  EmoLoom-2B turns small language models under 2B parameters into fast screening candidates for emotion classification and Valence-Arousal-Dominance prediction.<n>To ensure protocol-faithful and fair evaluation, we unify data loading, training, and inference under a single input-output contract.
  arXiv  Detail & Related papers  (2026-01-03T08:25:58Z)
• The Best of Both Worlds: Hybridizing Neural Operators and Solvers for Stable Long-Horizon Inference [0.0]
  ANCHOR is an online, instance-aware hybrid inference framework for stable long-horizon prediction of PDEs.<n>We show that ANCHOR reliably bounds long-horizon error growth, stabilizes extrapolative rollouts, and significantly improves robustness over standalone neural operators.
  arXiv  Detail & Related papers  (2025-12-22T18:17:28Z)
• Handling Imbalanced Pseudolabels for Vision-Language Models with Concept Alignment and Confusion-Aware Calibrated Margin [56.37346003683629]
  Adapting vision-language models (VLMs) to downstream tasks with pseudolabels has gained increasing attention.<n>A major obstacle is that the pseudolabels generated by VLMs tend to be imbalanced, leading to inferior performance.<n>We propose a novel framework incorporating concept alignment and confusion-aware margin mechanisms.
  arXiv  Detail & Related papers  (2025-05-04T10:24:34Z)

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.