arXiv Machine Learning
Jun 3
Attribution via Distributional Paths for Information Revelation
arXiv:2606. 03885v1 Announce Type: new Abstract: Feature attribution methods explain predictions by assigning importance scores to input features.
Decomposition of Evidence, Contradiction, and Fragility in Perturbation Responses
arXiv:2608. 12935v1 Announce Type: new Abstract: Perturbation methods explain model decisions by measuring prediction changes under altered inputs, but response magnitude tells us only how much a model reacts, not what that reaction means.
Related stories
arXiv Machine Learning
Aug 7
When Does Consensus Mean Correctness? Measuring the Agreement-Accuracy Coupling with Semantics-Preserving Re-Rendering
arXiv:2608. 05670v1 Announce Type: new Abstract: A model's agreement across perturbed inputs is used both as a label-free reliability signal and as a self-training target, on the premise that agreement tracks correctness.
arXiv AI
Jul 1
Surprise as a Signal for Plasticity and Metacognition
arXiv:2606. 31495v1 Announce Type: new Abstract: We study a single idea across two settings: that a prediction-error signal, computed by a small predictor over the latent space of a frozen encoder, can serve both as a gate on plasticity and as a substrate for metacognition.
arXiv AI
Jun 9
Strained Coherence: A Pre-Failure Signal in Coding Agent Execution Trajectories
arXiv:2606. 07889v1 Announce Type: cross Abstract: LLM-based coding agents sometimes acknowledge a problem in their own reasoning and then proceed anyway.
arXiv AI
Aug 3
Step-Level Visual Grounding Faithfulness Predicts Out-of-Distribution Generalization in Long-Horizon Vision-Language Models
arXiv:2603. 06828v2 Announce Type: replace-cross Abstract: We uncover a behavioral law of long-horizon vision-language models: models that maintain temporally grounded beliefs generalize better.
arXiv AI
5d ago
TRAPSBench: Vision-Language Models Encode but Fail to Express Epistemic Restraint
arXiv:2608. 13167v1 Announce Type: cross Abstract: When visual evidence is occluded or chaotic, models should abstain.
arXiv AI
Aug 12
Attention-Path Fragility as an Uncertainty Signal in Large Language Models
arXiv:2608. 11138v1 Announce Type: cross Abstract: We propose that a model's uncertainty about a token is reflected not only in the breadth of its output distribution but also in whether a confident prediction is \emph{fragile} under perturbation of its attention pathways.
arXiv AI
Aug 7
The Illusion of Visual Tool-Use: A Causal Audit of Thinking with Images
arXiv:2608. 06270v1 Announce Type: new Abstract: The "thinking-with-images" paradigm equips multimodal LLMs with active visual operations such as crop-and-zoom.
arXiv AI
Jul 3
Robust for the Wrong Reasons: The Representational Geometry of LLM Robustness to Science Skepticism
arXiv:2607. 01951v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly consulted on contested scientific questions, raising the concern that they will sycophantically retreat from established consensus when a user signals doubt -- drifting toward a false balance that treats settled science as one view among several.
arXiv Machine Learning
Jun 30
Internal-State Probes Read the Situation, Not the Action: Three Negative Results for Pre-Action Misalignment Monitoring
arXiv:2606. 30449v1 Announce Type: new Abstract: Probes on model internals could help monitor agentic systems if they identify harmful text or tool actions before those actions are generated.
arXiv Machine Learning
Jun 25
Perfect Detection, Failed Control: The Geometry of Knowing vs. Steering in Language Models
arXiv:2606. 24952v1 Announce Type: cross Abstract: A central aspiration of mechanistic interpretability is controllability: if we know where a behavior is represented in a model's activations, we should be able to modify it.
arXiv Machine Learning
Aug 5
Sensitivity, Causality, and Repair Dissociate: A Layer-Wise Analysis of Perturbation Robustness and Its Scaling
arXiv:2608. 03842v1 Announce Type: cross Abstract: When a language model fails on surface-perturbed input (typos, OCR noise, homophones), "which layer is responsible" has three natural operationalizations: where representations diverge most (sensitivity), where restoring clean activations recovers the prediction (causality), and where a small adapter can repair the damage (compensatory capacity) - and we show these three layer maps dissociate.