Reading Calibrated Uncertainty from Language Model Trajectories

• URL: http://arxiv.org/abs/2605.22864v1
• Date: Tue, 19 May 2026 19:24:29 GMT
• Title: Reading Calibrated Uncertainty from Language Model Trajectories
• Abstract summary: Methods that probe the model's internal activations feed raw hidden states into opaque snapshots and leave implicit the layer-wise trajectory by which a representation is formed.<n>We extract eleven scale-invariant geometric features, tracing the cumulative path of per-layer updates, and feed them to a sparse linear probe.<n>The probe outperforms MSP under selective abstention, with baseline scaling with up to 21 AURC points.
• Score: 46.663987199083245
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The maximum softmax probability (MSP) represents a default approach when evaluating uncertainty quantification for language model generation with structured output. Although cheap, it is often miscalibrated. Methods that probe the model's internal activations feed raw hidden states into opaque classifiers, reading activations as static snapshots and leaving implicit the layer-wise trajectory by which a representation is formed. Yet, similar endpoints can arise from very different paths, and how evidence accumulates, reinforces, or reverses across depth might reveal uncertainty that final probabilities obscure. We extract eleven scale-invariant geometric features, tracing the cumulative path of per-layer MLP updates, and feed them to a sparse linear probe. The probe outperforms MSP under selective abstention, with gains scaling with baseline miscalibration up to 21 AURC points. Because every feature has a closed-form geometric meaning, the probe's coefficients trace how and where along depth errors take shape -- which layers commit prematurely, which contradict the running state, where trajectories drift away from their endpoint.

Related papers

• Need We Teach Foundation Models What is a Generative Image? Gradient-Free Generative Artifact Detection via Analytic Spectral Adaptation [3.881211374324378]
  Adapting foundation models to detect generative artifacts via gradient-based updates compromises their intrinsic representations.<n>We propose a gradient-free methodology reframing detection from binary classification to an out-of-distribution (OOD) anomaly measurement problem.
  arXiv  Detail & Related papers  (2026-06-03T13:04:23Z)
• Active Timepoint Selection for Learning Measure-Valued Trajectories [55.49809248697036]
  Inferring continuous probability paths from snapshots is a fundamental challenge in domains like single-cell biology.<n>We introduce a framework which extends active experimentation to the space of sparse measures.<n>This yields an acquisition policy that iteratively selects measurement times to minimize uncertainty.
  arXiv  Detail & Related papers  (2026-05-28T22:22:35Z)
• Metric--Phase Fields: Decoupling Distance and Sign for Thin-Structure Reconstruction from Unoriented Point Clouds [67.30236447078187]
  We introduce Metric-Phase Fields (MPFs), a decoupled implicit representation that separates metric proximity from topological phase.<n>Experiments on both synthetic and scanned thin-shell and thin-plate shapes demonstrate that MPFs preserve thin and layered structures more faithfully than recent SDF-based methods.
  arXiv  Detail & Related papers  (2026-05-25T07:08:57Z)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• Inference Time Causal Probing in LLMs [26.69306706280534]
  In causal probing, an intervention modifies hidden states so that a property takes on a different value.<n>We propose Hidden-state Driven Margin Intervention (HDMI), a probe-free, gradient-based technique that directly steers hidden states.<n> HDMI consistently achieves higher reliability than prior methods on the LGD agreement corpus and the CausalGym benchmark.
  arXiv  Detail & Related papers  (2026-05-08T11:59:13Z)
• UnGAP: Uncertainty-Guided Affine Prompting for Real-Time Crack Segmentation [16.902729686979303]
  Uncertainty-Prompted Feature Modulator (UPFM) treats aleatoric uncertainty as an active visual prompt rather than a mere output.<n>UnGAP balances superior segmentation accuracy with real-time inference speed.
  arXiv  Detail & Related papers  (2026-05-04T09:18:57Z)
• Semantic Step Prediction: Multi-Step Latent Forecasting in LLM Reasoning Trajectories via Step Sampling [0.0]
  Semantic Tube Prediction (STP) leverages geometric representation to regularize hidden trajectories toward locally linear geodesics during fine-tuning.<n>We investigate whether the sampling position can further enhance the semantic structure of multi-step reasoning.<n>Our results identify sampling position as the critical variable in geometric regularization and establish multi-step latent prediction MSE as a new evaluation metric for this class of methods.
  arXiv  Detail & Related papers  (2026-04-20T16:19:02Z)
• GTS: Inference-Time Scaling of Latent Reasoning with a Learnable Gaussian Thought Sampler [54.10960908347221]
  We model latent thought exploration as conditional sampling from learnable densities and instantiate this idea as a Gaussian Thought Sampler (GTS)<n>GTS predicts context-dependent perturbation distributions over continuous reasoning states and is trained with GRPO-style policy optimization while keeping the backbone frozen.
  arXiv  Detail & Related papers  (2026-02-15T09:57:47Z)
• Phase-space entropy at acquisition reflects downstream learnability [54.4100065023873]
  We propose an acquisition-level scalar $S_mathcal B$ based on instrument-resolved phase space.<n>We show theoretically that (S_mathcal B) correctly identifies the phase-space coherence of periodic sampling.<n>$|S_mathcal B|$ consistently ranks sampling geometries and predicts downstream reconstruction/recognition difficulty emphwithout training.
  arXiv  Detail & Related papers  (2025-12-22T10:03:51Z)
• Probabilistic Skip Connections for Deterministic Uncertainty Quantification in Deep Neural Networks [2.186901738997927]
  Current deterministic methods often retrain networks with spectral normalization.<n>We propose using measures of neural collapse to identify an existing intermediate layer that is both sensitive and smooth.<n>We then fit a probabilistic model to the feature vector of this intermediate layer, which we call a probabilistic skip connection (PSC)
  arXiv  Detail & Related papers  (2025-01-08T20:12:33Z)
• Uncertainty-aware retinal layer segmentation in OCT through probabilistic signed distance functions [6.765624289092461]
  We present a new approach for uncertainty-aware retinal layer segmentation in Optical Coherence Tomography ( OCT) scans.<n>Our methodology refines the segmentation by predicting a signed distance function (SDF) that effectively parameterizes the retinal layer shape via level set.<n>This ensures a robust representation of the retinal layer even in the presence of ambiguous input, imaging noise, and unreliable segmentations.
  arXiv  Detail & Related papers  (2024-12-06T10:44:11Z)
• Sampling from Gaussian Process Posteriors using Stochastic Gradient Descent [43.097493761380186]
  gradient algorithms are an efficient method of approximately solving linear systems. We show that gradient descent produces accurate predictions, even in cases where it does not converge quickly to the optimum. Experimentally, gradient descent achieves state-of-the-art performance on sufficiently large-scale or ill-conditioned regression tasks.
  arXiv  Detail & Related papers  (2023-06-20T15:07:37Z)
• Robust Depth Completion with Uncertainty-Driven Loss Functions [60.9237639890582]
  We introduce uncertainty-driven loss functions to improve the robustness of depth completion and handle the uncertainty in depth completion. Our method has been tested on KITTI Depth Completion Benchmark and achieved the state-of-the-art robustness performance in terms of MAE, IMAE, and IRMSE metrics.
  arXiv  Detail & Related papers  (2021-12-15T05:22:34Z)

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.