An argument paper reframes LLM explainability as an embodied, situated practice based on Dourish and enactivist cognition, identifying ontological obstacles in internal explanations and advocating affordance-based designs.
Title resolution pending
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
support 1representative citing papers
Concept-based models can use controlled 'benign' information leakage to remain accurate and intervenable under real-world concept incompleteness by reframing their training objective.
Multimodal contrastive learning using multilinear products is fragile to single bad modalities, and a gated version improves top-1 retrieval accuracy on synthetic and real trimodal data.
Authors create a benchmark across discrete/continuous and static/dynamical systems and introduce the Causal Abstraction Error (CAE) metric that reliably distinguishes valid from invalid causal abstractions when it includes faithfulness testing.
Counterfactual baselines for Integrated Gradients yield more faithful and medically relevant attributions than standard baselines across three medical datasets.
RUBEN discovers minimal rule sets explaining RAG LLM outputs via novel pruning and applies them to evaluate LLM safety against adversarial injections.
citing papers explorer
-
Embodied Explainability and Ontological Obstacles: Why We Struggle to Explain the Answers of Large Language Models (LLMs)
An argument paper reframes LLM explainability as an embodied, situated practice based on Dourish and enactivist cognition, identifying ontological obstacles in internal explanations and advocating affordance-based designs.
-
In Defense of Information Leakage in Concept-based Models
Concept-based models can use controlled 'benign' information leakage to remain accurate and intervenable under real-world concept incompleteness by reframing their training objective.
-
Hidden in the Multiplicative Interaction: Uncovering Fragility in Multimodal Contrastive Learning
Multimodal contrastive learning using multilinear products is fragile to single bad modalities, and a gated version improves top-1 retrieval accuracy on synthetic and real trimodal data.
-
Validating Causal Abstraction Metrics on Simulated Complex Systems
Authors create a benchmark across discrete/continuous and static/dynamical systems and introduce the Causal Abstraction Error (CAE) metric that reliably distinguishes valid from invalid causal abstractions when it includes faithfulness testing.
-
On the notion of missingness for path attribution explainability methods in medical settings: Guiding the selection of medically meaningful baselines
Counterfactual baselines for Integrated Gradients yield more faithful and medically relevant attributions than standard baselines across three medical datasets.
-
RUBEN: Rule-Based Explanations for Retrieval-Augmented LLM Systems
RUBEN discovers minimal rule sets explaining RAG LLM outputs via novel pruning and applies them to evaluate LLM safety against adversarial injections.