Revisiting Self-Consistency from Dynamic Distributional Alignment Perspective on Answer Aggregation

• URL: http://arxiv.org/abs/2502.19830v2
• Date: Wed, 11 Jun 2025 07:11:03 GMT
• Title: Revisiting Self-Consistency from Dynamic Distributional Alignment Perspective on Answer Aggregation
• Abstract summary: Self-consistency improves reasoning by aggregating diverse samples.<n>We show that decoding temperature governs randomness and actively shapes the latent answer distribution.<n>We propose a confidence-driven mechanism that dynamically calibrates temperature.
• Score: 21.64693536216534
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-consistency improves reasoning by aggregating diverse stochastic samples, yet the dynamics behind its efficacy remain underexplored. We reframe self-consistency as a dynamic distributional alignment problem, revealing that decoding temperature not only governs sampling randomness but also actively shapes the latent answer distribution. Given that high temperatures require prohibitively large sample sizes to stabilize, while low temperatures risk amplifying biases, we propose a confidence-driven mechanism that dynamically calibrates temperature: sharpening the sampling distribution under uncertainty to align with high-probability modes, and promoting exploration when confidence is high. Experiments on mathematical reasoning tasks show this approach outperforms fixed-diversity baselines under limited samples, improving both average and best-case performance across varying initial temperatures without additional data or modules. This establishes self-consistency as a synchronization challenge between sampling dynamics and evolving answer distributions.

Related papers

• Diversify Diffusion with Temperature Sampling and Variance-Corrective Time Shifting [57.75750707482832]
  We introduce variance-corrective time shifting, a training-free fix that queries the network at a shifted timestep and scales the resulting score by $$, canceling the variance inflation while preserving the mode reweighting.<n>We demonstrate consistent gains at minimal cost to sample quality and condition fidelity across DiT, Stable Diffusion and Motion Diffusion models.
  arXiv  Detail & Related papers  (2026-07-12T17:32:29Z)
• Precise: SDE-Consistent Stochastic Sampling for RL Post-Training of Flow-Matching Models [56.67321805551389]
  Reinforcement learning (RL) has become an effective way to improve prompt alignment and perceptual quality in diffusion and flow-matching generators.<n>The sampler, controlling the exploration behavior and denoising dynamics, is thus part of the policy.<n>We propose a new sampler that balances effective exploration with stability.
  arXiv  Detail & Related papers  (2026-05-22T11:37:22Z)
• Inference-Time Attribute Distribution Alignment for Unconditional Diffusion [54.28942858023809]
  We formalize the inference-time attribute distributional alignment problem for pretrained unconditional diffusion models.<n>To address this, we cast inference-time attribute distributional alignment as an optimal control problem over the reverse diffusion process.<n>We solve for the perturbations using an optimal-control-based algorithm to optimize a differentiable distribution-matching objective.
  arXiv  Detail & Related papers  (2026-05-08T09:02:47Z)
• Improving Semantic Uncertainty Quantification in Language Model Question-Answering via Token-Level Temperature Scaling [55.29156356409814]
  We show that current approaches, particularly fixed-temperatures, produce systematically miscalibrated and poorly discriminative semantic confidence distributions.<n>We demonstrate that optimising a single temperature, which, we argue, provides a suitable inductive bias, is a surprisingly simple yet effective solution.
  arXiv  Detail & Related papers  (2026-04-08T15:00:32Z)
• A Generative Sampler for distributions with possible discrete parameter based on Reversibility [9.349846971147256]
  We propose a unified, target-gradient-free generative sampling framework applicable across diverse state spaces.<n>We minimize the Maximum Mean Discrepancy (MMD) between the joint distributions of forward and backward Markov trajectories.<n> Experiments show that our framework accurately reproduces thermodynamic observables and captures mode-switching behavior across all regimes.
  arXiv  Detail & Related papers  (2026-03-10T06:32:51Z)
• Drift No More? Context Equilibria in Multi-Turn LLM Interactions [58.69551510148673]
  contexts drift is the gradual divergence of a model's outputs from goal-consistent behavior across turns.<n>Unlike single-turn errors, drift unfolds temporally and is poorly captured by static evaluation metrics.<n>We show that multi-turn drift can be understood as a controllable equilibrium phenomenon rather than as inevitable decay.
  arXiv  Detail & Related papers  (2025-10-09T04:48:49Z)
• Semantically-Guided Inference for Conditional Diffusion Models: Enhancing Covariate Consistency in Time Series Forecasting [6.716179859091235]
  SemGuide is a plug-and-play, inference-time method that enhances covariate consistency in conditional diffusion models.<n>Our approach introduces a scoring network to assess the semantic alignment between intermediate diffusion states and future covariates.
  arXiv  Detail & Related papers  (2025-08-03T14:04:04Z)
• Advancing Reliable Test-Time Adaptation of Vision-Language Models under Visual Variations [67.35596444651037]
  Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable.<n>We propose a Reliable Test-time Adaptation (ReTA) method that enhances reliability from two perspectives.
  arXiv  Detail & Related papers  (2025-07-13T05:37:33Z)
• Enhancing Gradient-based Discrete Sampling via Parallel Tempering [8.195708231156546]
  A gradient-based discrete sampler is susceptible to getting trapped in local minima in high-dimensional, multimodal discrete distributions.<n>We develop a discrete Langevin proposal that combines parallel tempering, also known as replica exchange, with the discrete Langevin proposal.<n>We show that our algorithm converges non-asymptotically to the target energy and exhibits faster mixing compared to a single chain.
  arXiv  Detail & Related papers  (2025-02-26T15:51:15Z)
• Federated Learning with Sample-level Client Drift Mitigation [15.248811557566128]
  Federated Learning suffers from severe performance degradation due to data heterogeneity among clients. We propose FedBSS that first mitigates the heterogeneity issue in a sample-level manner. We also achieved effective results on feature distribution and noise label dataset setting.
  arXiv  Detail & Related papers  (2025-01-20T09:44:07Z)
• Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers [49.97755400231656]
  We present the first performance guarantee with explicit dimensional general score-mismatched diffusion samplers. We show that score mismatches result in an distributional bias between the target and sampling distributions, proportional to the accumulated mismatch between the target and training distributions. This result can be directly applied to zero-shot conditional samplers for any conditional model, irrespective of measurement noise.
  arXiv  Detail & Related papers  (2024-10-17T16:42:12Z)
• Efficient Sampling for Data-Driven Frequency Stability Constraint via Forward-Mode Automatic Differentiation [5.603382086370097]
  We propose a gradient-based data generation method via forward-mode automatic differentiation. In this method, the original dynamic system is augmented with new states that represent the dynamic of sensitivities of the original states. We demonstrate the superior performance of the proposed sampling algorithm, compared with the unrolling differentiation and finite difference.
  arXiv  Detail & Related papers  (2024-07-21T03:50:11Z)
• Extreme Miscalibration and the Illusion of Adversarial Robustness [66.29268991629085]
  Adversarial Training is often used to increase model robustness. We show that this observed gain in robustness is an illusion of robustness (IOR) We urge the NLP community to incorporate test-time temperature scaling into their robustness evaluations.
  arXiv  Detail & Related papers  (2024-02-27T13:49:12Z)
• Selective Learning: Towards Robust Calibration with Dynamic Regularization [79.92633587914659]
  Miscalibration in deep learning refers to there is a discrepancy between the predicted confidence and performance. We introduce Dynamic Regularization (DReg) which aims to learn what should be learned during training thereby circumventing the confidence adjusting trade-off.
  arXiv  Detail & Related papers  (2024-02-13T11:25:20Z)
• Benchmarking Autoregressive Conditional Diffusion Models for Turbulent Flow Simulation [26.520247496906492]
  In this work, we analyze if fully data-driven fluid solvers that utilize an autoregressive rollout based on conditional diffusion models are a viable option to address this challenge.<n>To quantitatively and qualitatively benchmark the performance of various flow prediction approaches, three challenging 2D scenarios including incompressible and transonic flows, as well as isotropic turbulence are employed.<n>We find that even simple diffusion-based approaches can outperform multiple established flow prediction methods in terms of accuracy and temporal stability, while being on par with state-of-the-art stabilization techniques like unrolling at training time.
  arXiv  Detail & Related papers  (2023-09-04T18:01:42Z)
• Dynamically Scaled Temperature in Self-Supervised Contrastive Learning [11.133502139934437]
  We focus on improving the performance of InfoNCE loss in self-supervised learning by proposing a novel cosine similarity dependent temperature scaling function. Experimental evidence shows that the proposed framework outperforms the contrastive loss-based SSL algorithms.
  arXiv  Detail & Related papers  (2023-08-02T13:31:41Z)
• Breaking the Spurious Causality of Conditional Generation via Fairness Intervention with Corrective Sampling [77.15766509677348]
  Conditional generative models often inherit spurious correlations from the training dataset. This can result in label-conditional distributions that are imbalanced with respect to another latent attribute. We propose a general two-step strategy to mitigate this issue.
  arXiv  Detail & Related papers  (2022-12-05T08:09:33Z)
• Inference in conditioned dynamics through causality restoration [0.0]
  We propose an alternative method to produce independent samples from a conditioned distribution. The method learns the parameters of a generalized dynamical model. We discuss an important application of the method, namely the problem of epidemic risk assessment from (imperfect) clinical tests.
  arXiv  Detail & Related papers  (2022-10-18T21:58:58Z)
• Robust Calibration with Multi-domain Temperature Scaling [86.07299013396059]
  We develop a systematic calibration model to handle distribution shifts by leveraging data from multiple domains. Our proposed method -- multi-domain temperature scaling -- uses the robustness in the domains to improve calibration under distribution shift.
  arXiv  Detail & Related papers  (2022-06-06T17:32:12Z)
• Trustworthy Multimodal Regression with Mixture of Normal-inverse Gamma Distributions [91.63716984911278]
  We introduce a novel Mixture of Normal-Inverse Gamma distributions (MoNIG) algorithm, which efficiently estimates uncertainty in principle for adaptive integration of different modalities and produces a trustworthy regression result. Experimental results on both synthetic and different real-world data demonstrate the effectiveness and trustworthiness of our method on various multimodal regression tasks.
  arXiv  Detail & Related papers  (2021-11-11T14:28:12Z)

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.