Uncertainty and verification
24 notes, newest first. Includes the vault tags uncertainty, verification.
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
The preprint evaluates epistemic humility through three trajectory-level behaviors: Identify, Solve, and Escalate. It tests controlled and naturally occurring conflicts between parametric knowledge and retrieved…
Searching for "Harmful Refusal": A Psychometric Audit of an AI Safety Benchmark
The preprint audits whether harmful refusal is a single measurable attribute in HELM Safety. It reports saturation in three candidate datasets and uses multidimensional item-response theory plus differential item…
Cogentic: Multi-Agent Orchestration for Automated Proof Discovery
The preprint describes an orchestrator that assigns independent provers to competing proof directions, uses specialized adversarial verification components, and promotes confirmed intermediate results into a persistent…
HARISSA: Inference-Time Self-Checks for Efficient and Safe Local Language Model Deployment
The preprint uses hidden states before generation and at the end of an answer to predict correctness for local-model deployment. HARISSA fine-tunes both states and uses a single policy to choose among increasingly…
NNV3: Expanding Neural Network Verification to New Architectures and Domains
The preprint presents NNV3, a MATLAB framework for formal verification of neural networks. It describes ModelStar for weight perturbation, VolumeStar for video and 3D inputs, GraphStar for graph neural networks…
JEV-as-a-Judge: Accept When Confident, Escalate When Unsure
The preprint evaluates a decision-only judge as an economical first pass and escalates low-confidence cases to stronger evaluation. It compares the judge with 16 generative and reward-model judges using blinded human…
Pinocchio: Fast Uncertainty Estimates for Black-Box Language Models
Pinocchio is an external calibrator for estimating response correctness from black-box language-model APIs without logits, weights, internal states, or fine-tuning access. The preprint reports a single-forward-pass…
How Do Agent Harnesses Create Value? Planning Information and Release Control in Stateful LLM Agents
Across 265 matched cells in Retail experiments and an Airline pilot in τ²-bench, the paper compares task-specific plans with shuffled text, and evaluates a read-only terminal verifier. It reports a 7.17-point…
Refuse, Decompose, Refresh: A Claim-Safe Protocol for Closed-Loop AI Evaluation
The protocol proposes three actions: refuse when a clean reference stream or matched runtime comparison lacks support; decompose execution, false admission, and structural hypotheses instead of collapsing them into…
PosteriorBench: From Point Estimates to Posterior Matching in Evaluating Generative Inverse Solvers
PosteriorBench evaluates whether generative inverse solvers recover full solution distributions rather than only one plausible reconstruction. It covers four physics-based inverse problems and uses reference posteriors…
In-Context Robot Learning with VLM Agents
The paper introduces GPT-Policy, a framework combining a context compiler, a vision-language model that proposes robot-tool actions, and a constrained controller that verifies and executes actions. The authors report…
ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation
ENCP rescales conformal-prediction scores for variable-length, dependent navigation episodes and calibrates one maximum score per episode. The abstract reports coverage evaluations across four navigation policies and…
When Should LLMs Abstain? Chain-of-Self-Questioning for Selective Risk Control
The paper proposes Chain-of-Self-Questioning, a prompt-only framework that makes answer commitment conditional on an explicit assessment of whether the model has enough information. The abstract reports evaluations on…
Falsifiable Commitment Planning for Self-Correcting Web Agents
Researchers propose a new framework called FCPAgent for web agents to ensure they stay on track during long-term planning. The framework represents each plan step as a Falsifiable Commitment Unit (FCU) that includes…
Partially Correlated Verifier Cascades in LLM Harnesses: Concave Log-Odds, Polynomial Reliability, and Blind-Spot Ceilings
The text presents a new theory for partially correlated verifier cascades in LLM harnesses (a system used to verify candidate answers). The theory provides insights into the growth and decay of failure rates, the…
Evidence-Grounded Verified Agentic Reasoning: A Path Toward Eliminating LLM Hallucination in Empirical Inference via Tool-Attested Kernel Proofs
The paper introduces EG-VAR, a tool-calling architecture that aims to make the reasoning of a large language model (LLM) more trustworthy by verifying all outputs through tool-attestation axioms and kernel checks. The…
Physics-Audited Agentic Discovery in Scientific Machine Learning
The paper introduces Physics-Audited Agentic Scientific Machine Learning (PA-SciML), a new workflow that verifies the accuracy of surrogate models discovered by large language model agents in scientific machine…
Estimating Uncertainty from Reasoning: A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLMs
The researchers evaluate the performance of uncertainty estimation methods across 22 languages in high-, mid-, and low-resource settings using two human-curated question and answer datasets. They compare open and…
Harnessing Code Agents for Automatic Software Verification
The authors propose a new method for automatic software verification using large language models (LLMs) in the form of 'code agents'. Instead of wireting a fixed proof strategy, they allow the code agent to choose its…
LLM-as-a-Verifier: A General-Purpose Verification Framework
The paper introduces LLM-as-a-Verifier, a general-purpose verification framework that provides fine-grained feedback to improve the accuracy of agentic tasks. The framework scales across multiple dimensions for better…
LLM-as-a-Verifier: A General-Purpose Verification Framework
In this research, a general-purpose verification framework called LLM-as-a-Verifier is introduced. This framework provides detailed feedback for agentic tasks without requiring additional training. The framework…
Theoria: Rewrite-Acceptability Verification over Informal Reasoning States
Theoria is a verification architecture proposed by Ben Slivinski and Michael Saldivar that aims to close the gap between formal proof assistants and scalar LLM judges in determining when an AI system's answer can be…
AxDafny: Agentic Verified Code Generation in Dafny
Researchers present AxDafny, a framework for generating executable code and proof artifacts in Dafny using a verifier-guided repair method. The study uses a benchmark called LiveCodeBench-Pro-Dafny with 250 programming…
Bayesian control for coding agents
The paper proposes a Bayesian controller for coding agents that make decisions based on a belief over candidate correctness, rather than fixed rules. This method is found to be most useful when verification is costly…
