Cover Image

Measuring Black-Box Confidence via Reasoning Trajectories: Geometry, Coverage, and Verbalization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Martell, Marc Boubnovski, Stoisser, Josefa Lia, Märtens, Kaspar, Yu, Jialin, Kitchen, Robert, Torr, Philip, Ferkinghoff-Borg, Jesper
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MechPert: Mechanistic Consensus as an Inductive Bias for Unseen Perturbation Prediction
by: Martell, Marc Boubnovski, et al.
Published: (2026)

Ambig-DS: A Benchmark for Task-Framing Ambiguity in Data-Science Agents
by: Stoisser, Josefa Lia, et al.
Published: (2026)

Towards Agents That Know When They Don't Know: Uncertainty as a Control Signal for Structured Reasoning
by: Stoisser, Josefa Lia, et al.
Published: (2025)

Sparks of Tabular Reasoning via Text2SQL Reinforcement Learning
by: Stoisser, Josefa Lia, et al.
Published: (2025)

Towards Label-Free Biological Reasoning Synthetic Dataset Creation via Uncertainty Filtering
by: Stoisser, Josefa Lia, et al.
Published: (2025)

SynthPert: Enhancing LLM Biological Reasoning via Synthetic Reasoning Traces for Cellular Perturbation Prediction
by: Phillips, Lawrence, et al.
Published: (2025)

STRuCT-LLM: Unifying Tabular and Graph Reasoning with Reinforcement Learning for Semantic Parsing
by: Stoisser, Josefa Lia, et al.
Published: (2025)

Query, Don't Train: Privacy-Preserving Tabular Prediction from EHR Data via SQL Queries
by: Stoisser, Josefa Lia, et al.
Published: (2025)

Batched Energy-Entropy acquisition for Bayesian Optimization
by: Teufel, Felix, et al.
Published: (2024)

Zero-shot protein stability prediction by inverse folding models: a free energy interpretation
by: Frellsen, Jes, et al.
Published: (2025)

Disentangling shared and private latent factors in multimodal Variational Autoencoders
by: Märtens, Kaspar, et al.
Published: (2024)

On Verbalized Confidence Scores for LLMs
by: Yang, Daniel, et al.
Published: (2024)

Are LLM Decisions Faithful to Verbal Confidence?
by: Wang, Jiawei, et al.
Published: (2026)

Direct Confidence Alignment: Aligning Verbalized Confidence with Internal Confidence In Large Language Models
by: Zhang, Glenn, et al.
Published: (2025)

Verbalized Confidence Triggers Self-Verification: Emergent Behavior Without Explicit Reasoning Supervision
by: Jang, Chaeyun, et al.
Published: (2025)

ADVICE: Answer-Dependent Verbalized Confidence Estimation
by: Seo, Ki Jung, et al.
Published: (2025)

How do LLMs Compute Verbal Confidence
by: Kumaran, Dharshan, et al.
Published: (2026)

On the Robustness of Verbal Confidence of LLMs in Adversarial Attacks
by: Obadinma, Stephen, et al.
Published: (2025)

Calibrating Verbalized Confidence with Self-Generated Distractors
by: Wang, Victor, et al.
Published: (2025)

From Black-Box Confidence to Measurable Trust in Clinical AI: A Framework for Evidence, Supervision, and Staged Autonomy
by: Zabolotnii, Serhii, et al.
Published: (2026)

LiBOG: Lifelong Learning for Black-Box Optimizer Generation
by: Pei, Jiyuan, et al.
Published: (2025)

A Leakage Bound for Confidence Sets after Black-Box Selection
by: Banerjee, Sayantan
Published: (2026)

PCS: Perceived Confidence Scoring of Black Box LLMs with Metamorphic Relations
by: Salimian, Sina, et al.
Published: (2025)

Large Language Model Confidence Estimation via Black-Box Access
by: Pedapati, Tejaswini, et al.
Published: (2024)

Time of day of cardiac surgery and postoperative outcomes: a reply
by: Gareth Kitchen
Published: (2026)

Breaking the Black-Box: Confidence-Guided Model Inversion Attack for Distribution Shift
by: Liu, Xinhao, et al.
Published: (2024)

Realizing an Atomtronic AQUID in a Rotating-Box Potential
by: Görg, Kaspar, et al.
Published: (2025)

Modeling Gene Expression Distributional Shifts for Unseen Genetic Perturbations
by: Ramakrishnan, Kalyan, et al.
Published: (2025)

"Moralized" Multi-Step Jailbreak Prompts: Black-Box Testing of Guardrails in Large Language Models for Verbal Attacks
by: Wang, Libo
Published: (2024)

Are Large Language Models More Honest in Their Probabilistic or Verbalized Confidence?
by: Ni, Shiyu, et al.
Published: (2024)

Influential Training Data Retrieval for Explaining Verbalized Confidence of LLMs
by: Xia, Yuxi, et al.
Published: (2026)

Verbal and Numeric Eyewitness Confidence Differentially Affect Decision‐Making
by: Pia Pennekamp
Published: (2025)

Certifiable Black-Box Attacks with Randomized Adversarial Examples: Breaking Defenses with Provable Confidence
by: Hong, Hanbin, et al.
Published: (2023)

SLIM: Stealthy Low-Coverage Black-Box Watermarking via Latent-Space Confusion Zones
by: Wu, Hengyu, et al.
Published: (2026)

PUH Theorem 214 — Black Hole Information Preservation in Planck Cores via Polariton Bound States, and the Triple Isolation of the Matter and Antimatter Universes Across the Rebound Throat: A Resolution of the Hawking Information Paradox Within the Photonic Universe Hypothesis
by: Martell, Brian
Published: (2026)

Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs
by: Zhao, Tianyi, et al.
Published: (2026)

ConfTuner: Training Large Language Models to Express Their Confidence Verbally
by: Li, Yibo, et al.
Published: (2025)

ORCE: Order-Aware Alignment of Verbalized Confidence in Large Language Models
by: Li, Chen, et al.
Published: (2026)

Conformal Sets in Multiple-Choice Question Answering under Black-Box Settings with Provable Coverage Guarantees
by: Yang, Guang, et al.
Published: (2025)

Frozen/Thawed Samples Can Replace Fresh Samples for Assignment of ISI to Secondary Thromboplastin Standards for Multiple Reagent/Instrument Combinations: Data to Support Possible Revision of WHO Guidelines
by: Matthew Kitchen, et al.
Published: (2024)