The Geometric Nature and a Free Proxy for Flow-Matching Uncertainty

• URL: http://arxiv.org/abs/2607.27933v1
• Date: Thu, 30 Jul 2026 09:44:46 GMT
• Title: The Geometric Nature and a Free Proxy for Flow-Matching Uncertainty
• Abstract summary: Flow matching (FM) has become a popular action head paradigm for modern embodied models.<n>We show that uncertainty manifests as deviation from an ideal affine-isotropic contraction field.<n>We introduce denoising acceleration ($mathrmaccel$), a cost-free uncertainty proxy that measures the bending of the trajectory from a single forward pass.
• Score: 40.15880250907137
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Flow matching (FM) has become a popular action head paradigm for modern embodied models. However, as a conditional generative model, it does not explicitly expose its inherent uncertainty, producing faulty action chunks even when it misinterprets the scene or encounters out-of-distribution (OOD) inputs. Therefore, determining when an FM-generated action can be trusted is essential for safe deployment, yet existing uncertainty estimation methods on real-time control suffer from several issues: extra training budget, high computational overhead, and low generalization ability. In this work, we provide a geometric interpretation of FM uncertainty in the velocity field, showing that uncertainty manifests as deviation from an ideal affine-isotropic contraction field. Building on this observation, we introduce denoising acceleration ($\mathrm{accel}$), a highly-generalizable and cost-free uncertainty proxy that measures the bending of the denoising trajectory from a single forward pass, without additional model evaluations, training, or resampling. We theoretically and empirically demonstrate that $\mathrm{accel}$ is a faithful proxy for FM uncertainty and further test its utility in online failure detection. Results show that $\mathrm{accel}$ identifies failing rollouts well before termination, matching or even outperforming costly resampling- and training-based baselines across settings under realistic deployment budget. Code and demos available at: https://github.com/rrrrrrzy/fm-geometry.

Related papers

• Factor-Informed Uncertainty Distillation for Gaze Estimation [3.7508723681775606]
  We propose Factor-Informed Uncertainty Distillation (FIUD) to align uncertainty with interpretable image-quality failure modes.<n>A gradient-boosting teacher predicts expected gaze error from factors such as illumination, sharpness, eye visibility and symmetry.<n>A neural student distills these signals via curriculum learning and ranking supervision into a lightweight single-pass uncertainty head.
  arXiv  Detail & Related papers  (2026-07-22T12:20:40Z)
• Exploring Zero-Shot Foundation Models for Multivariate Time Series Anomaly Detection [4.468174618435543]
  MTSAD is essential for reliability and safety in domains such as industrial process monitoring and financial risk management.<n>Foundation Models (FMs) are pre-trained on broad data with strong zero-shot generalization.<n>We investigate the zero-shot application of a FM to industrial MTSAD on the Secure Water Treatment (SWaT) benchmark.
  arXiv  Detail & Related papers  (2026-07-14T07:32:08Z)
• Expectations vs. Realities: The Cost of MSE-Optimal Forecasting Under Conditional Uncertainty [1.2489632787815885]
  Multi-step time series forecasting (MSF) is commonly evaluated using pointwise error metrics such as mean squared error (MSE)<n>We show that this can be misleading under conditional uncertainty, where the conditional expectation becomes unrepresentative of typical realized values at longer horizons.<n>This establishes a fundamental, model-agnostic trade-off between point accuracy and marginal realism in MSF evaluation.
  arXiv  Detail & Related papers  (2026-06-03T01:50:32Z)
• LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification Generation [75.05397479715576]
  Large language models (LLMs) and agents have shown promising progress, but their true capabilities and failure modes remain unclear.<n>We present the first systematic and contamination-aware study of LLM- and agent-based formal specification generation for C programs.
  arXiv  Detail & Related papers  (2026-05-02T11:31:33Z)
• U$^{2}$Flow: Uncertainty-Aware Unsupervised Optical Flow Estimation [18.3748504668909]
  Unsupervised optical flow methods typically lack reliable uncertainty estimation, limiting their robustness and interpretability.<n>We propose U$2$Flow, the first recurrent unsupervised framework that jointly estimates optical flow and per-pixel uncertainty.
  arXiv  Detail & Related papers  (2026-04-11T06:40:22Z)
• Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol [69.11739400975445]
  We introduce the first theoretical framework for analyzing error accumulation in Model Context Protocol (MCP) agents.<n>We show that cumulative distortion exhibits linear growth and high-probability deviations bounded by $O(sqrtT)$.<n>Key findings include: semantic weighting reduces distortion by 80%, and periodic re-grounding approximately every 9 steps suffices for error control.
  arXiv  Detail & Related papers  (2026-02-10T21:08:53Z)
• ProbFM: Probabilistic Time Series Foundation Model with Uncertainty Decomposition [0.12489632787815884]
  Time Series Foundation Models (TSFMs) have emerged as a promising approach for zero-shot financial forecasting.<n>Current approaches either rely on restrictive distributional assumptions, conflate different sources of uncertainty, or lack principled calibration mechanisms.<n>We present a novel transformer-based probabilistic framework, ProbFM, that leverages Deep Evidential Regression (DER) to provide principled uncertainty quantification.
  arXiv  Detail & Related papers  (2026-01-15T17:02:06Z)
• The Illusion of Certainty: Uncertainty quantification for LLMs fails under ambiguity [48.899855816199484]
  We introduce MAQA* and AmbigQA*, the first ambiguous question-answering (QA) datasets equipped with ground-truth answer distributions.<n>We show that predictive-distribution and ensemble-based estimators are fundamentally limited under ambiguity.
  arXiv  Detail & Related papers  (2025-11-06T14:46:35Z)
• The Implicit Delta Method [61.36121543728134]
  In this paper, we propose an alternative, the implicit delta method, which works by infinitesimally regularizing the training loss of uncertainty. We show that the change in the evaluation due to regularization is consistent for the variance of the evaluation estimator, even when the infinitesimal change is approximated by a finite difference.
  arXiv  Detail & Related papers  (2022-11-11T19:34:17Z)
• Approaching Neural Network Uncertainty Realism [53.308409014122816]
  Quantifying or at least upper-bounding uncertainties is vital for safety-critical systems such as autonomous vehicles. We evaluate uncertainty realism -- a strict quality criterion -- with a Mahalanobis distance-based statistical test. We adopt it to the automotive domain and show that it significantly improves uncertainty realism compared to a plain encoder-decoder model.
  arXiv  Detail & Related papers  (2021-01-08T11:56:12Z)
• Trust but Verify: Assigning Prediction Credibility by Counterfactual Constrained Learning [123.3472310767721]
  Prediction credibility measures are fundamental in statistics and machine learning. These measures should account for the wide variety of models used in practice. The framework developed in this work expresses the credibility as a risk-fit trade-off.
  arXiv  Detail & Related papers  (2020-11-24T19:52:38Z)
• Deep Learning based Uncertainty Decomposition for Real-time Control [9.067368638784355]
  We propose a novel method for detecting the absence of training data using deep learning. We show its advantages over existing approaches on synthetic and real-world datasets. We further demonstrate the practicality of this uncertainty estimate in deploying online data-efficient control on a simulated quadcopter.
  arXiv  Detail & Related papers  (2020-10-06T10:46:27Z)

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.