Explainable Artificial Intelligence (XAI)

Derniers papiers

• Testing Frontier Large Language Models' Physics Literacy in Parallel Physical Worlds

  Dong Zhang · 2 juillet 2026

  Current large-language-model (LLM) physics benchmarks are usually scored by answer accuracy, which cannot distinguish genuine reasoning from recall of familiar problem patterns and reveals little about where a model's reasoning breaks down. We introduce an auditable four-stage diagnostic that evalua…

• Explaining Tabular Foundation Model Differences Through Meta-Features

  Markus Herre, Andrej Tschalzev, Sascha Marton, Christian Bartelt · 2 juillet 2026

  With the rise of tabular foundation models alongside traditional models still performing well on many tasks, choosing the right model for a tabular dataset remains difficult. We investigate whether dataset meta-features can explain performance gaps between model families on tabular prediction tasks.…

• Verbosity Tradeoffs and the Impact of Scale on the Faithfulness of LLM Self-Explanations

  Noah Y. Siegel, Nicolas Heess, Maria Perez-Ortiz, Oana-Maria Camburu · 2 juillet 2026

  When asked to explain their decisions, LLMs can often give explanations which sound plausible to humans. But are these explanations faithful, i.e. do they convey the factors actually responsible for the decision? In this work, we analyse counterfactual faithfulness across 75 models from 13 families.…

• The Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology

  Andrzej Szablewski, Gabriel Konar-Steenberg, Raffaello Fornasiere, Nikita Menon, Stefan Heimersheim · 2 juillet 2026

  Model organisms (MOs) - language models trained to exhibit undesired or unnatural behaviours - are frequently used as testbeds for evaluating white-box interpretability techniques. Current MOs are typically constructed via post-hoc supervised fine-tuning (SFT) on behavioural transcripts or synthetic…

• Distill to Detect: Exposing Stealth Biases in LLMs through Cartridge Distillation

  Shayan Talaei, Abhinav Chinta, Devvrit Khatri, Amin Karbasi, Azalia Mirhoseini, Amin Saberi · 2 juillet 2026

  Language models deployed in high-stakes roles can potentially favor certain entities, brands, or viewpoints, steering user decisions at scale. Such preferential biases can be introduced by any actor in the model's supply chain and are most dangerous when the model reveals its preference only on the …

• Shapley in Context: Explaining Financial Language with Domain Expertise

  Dangxing Chen, Pengzhan Guo · 2 juillet 2026

  In recent years, large language models have achieved remarkable success and have seen growing adoption in financial applications. At the same time, explainability remains critical in finance, a domain characterized by high stakes and strict regulatory requirements. Although numerous methods have bee…

• Testing Frontier Large Language Models' Physics Literacy in Parallel Physical Worlds

  Dong Zhang · 2 juillet 2026

  Current large-language-model (LLM) physics benchmarks are usually scored by answer accuracy, which cannot distinguish genuine reasoning from recall of familiar problem patterns and reveals little about where a model's reasoning breaks down. We introduce an auditable four-stage diagnostic that evalua…

• Harnessing the Latent Space: From Steering Vectors to Model Calibrators for Control and Trust

  Nishant Subramani · 2 juillet 2026

  Language models have changed from unreliable text generators to highly-capable large models with trillions of parameters. Capability increases come hand-in-hand with increases in scale, making understanding the internal representations of models more challenging. Since millions of users increasing r…

• Representation as a Bottleneck for Mechanistic Interpretability: The Manifestation Unit Protocol

  Hussein Chouman, Wataru Sasaki, Tomokazu Matsui, Hirohiko Suwa, Keiichi Yasumoto · 2 juillet 2026

  Mechanistic interpretability has produced a rich inventory of component-level analyses that characterise what neural-network components encode and how they interact. Their outputs, however, are not easily reusable: selectivity tables, circuit diagrams, and feature lists remain locked in per-study no…

• Explainability in mulimodal deep transformation models for stroke outcome prediction

  Lisa Herzog, Jonas Br\"andli, Maurice Schneeberger, Loran Avci, Nordin Dari, Martin H\"ansel, Hakim Baazaoui, Pascal B\"uhler, Susanne Wegener, Beate Sick · 2 juillet 2026

  Multimodal prediction models based on imaging and clinical data are increasingly used for clinical decision support, yet their interpretability remains limited. We present multimodal Deep Transformation Models (DTMs) combining statistical approaches and neural networks to achieve strong predictive p…

• TreeAgent: A Generalizable Multi-Agent Framework for Automated Bias Labeling in Forestry via Compiled Expert Rules and Vision-Language Models

  Shiyi Chen, Nicholas Saban, Collin Hargreaves, Huiqi Wang · 1 juillet 2026

  Human-labeled data are widely used as reference annotations in ML, despite known variability across annotators in many expert-driven domains. In addition, expert annotation is slow, inconsistent, and remains a major bottleneck for scaling tasks like tree height bias classification in forestry remote…

• Scientific Explanations in Health Sciences: Causality, Trust, and Epistemic Adequacy

  Martina Mattioli, Marcello Pelillo · 1 juillet 2026

  Medical Artificial Intelligence (AI) is widely expected to transform clinical practice, yet the decision-making processes of many Machine Learning (ML) models remain opaque. Explainability has been advanced as a partial remedy to clarify why AI generates predictions, particularly in high-stakes cont…

• Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs

  Gabrielle Kaili-May Liu, Avi Caciularu, Gal Yona, Idan Szpektor, Arman Cohan · 1 juillet 2026

  Metacognition is a critical component of intelligence that describes the ability to monitor and regulate one's own cognitive processes. Yet LLMs exhibit systemic deficiencies in key metacognitive faculties: they hallucinate with high confidence, fail to recognize knowledge boundaries, and misreprese…

• DISCOVER: A Solver for Distributional Counterfactual Explanations

  Yikai Gu, Lele Cao, Bo Zhao, Lei Lei, Lei You · 1 juillet 2026

  Counterfactual explanations (CE) explain model decisions by identifying input modifications that lead to different predictions. Most existing methods operate at the instance level. Distributional Counterfactual Explanations (DCE) extend this setting by optimizing an optimal transport objective that …

• Deductive Logic in Language Models: Horizontal vs Vertical Reasoning

  Davide Maltoni, Matteo Ferrara · 1 juillet 2026

  Recent language models exhibit significant logical reasoning abilities, yet the mechanisms supporting deductive inference remain poorly understood. This paper studies small transformer-based language models trained from scratch on multi-step deductive tasks, focusing on the distinction between horiz…

• Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision

  Zifan Carl Guo, Laura Ruis, Jacob Andreas, Belinda Z. Li · 1 juillet 2026

  When does training language models (LMs) to generate explanations of their predictions yield faithful introspection, rather than superficial imitation? We study LMs trained to explain which features of their inputs influenced their behavior, using models' counterfactual behavior on modified inputs a…

• Calibrating the Evaluator: Does Probability Calibration Mitigate Preference Coupling in LLM Agent Feedback Loops?

  Zewen Liu · 1 juillet 2026

  When large language model (LLM) agents adapt their behavior through evaluator feedback, systematic evaluator biases propagate into the agent's learned strategy distribution - a phenomenon termed evaluator preference coupling. Prior work has documented this coupling and established a diagnostic frame…

• Multistage Defer Trees for Hybrid Interpretability: If at First You Can't Succeed, Tree Again

  Zakk Heile, Hayden McTavish, Margo Seltzer, Cynthia Rudin · 1 juillet 2026

  Recent work has shown that well-optimized individual decision trees can match complex black box models in some settings, primarily in noisy domains. For the remaining settings, however, complex ensembled compositions of trees often achieve higher accuracy at the cost of interpretability, leaving pra…

• Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?

  Philippe Chlenski, Zachariah Carmichael, Ayush Warikoo, Chia-Tse Shao, Yingxiao Ye, Aobo Yang, Vivek Miglani, Nehal Bandi · 1 juillet 2026

  Mechanistic interpretability (MI) requires full access to model internals, yet the APIs for most widely deployed language models at best expose log-probabilities over output tokens. This creates a surrogate problem: when do measurements made on open models allow us to make claims about a closed mode…

• Neuro-Bayesian-Symbolic Residual Attention Shallow Network: Explainable Deep Learning for Cybersecurity Risk Assessment

  Nicolaie Popescu-Bodorin, Madeleine Togher · 1 juillet 2026

  We introduce the Neuro-Bayesian-Symbolic Residual Attention Shallow Network (NBS-RASN), a hybrid neural architecture for explainable cybersecurity risk assessment in open-source ecosystems. Unlike deep models that trade interpretability for accuracy, our shallow network encodes domain knowledge, cau…

• Compositional Concept-Based Neuron-Level Interpretability for Deep Reinforcement Learning

  Zeyu Jiang, Hai Huang, Xingquan Zuo · 1 juillet 2026

  Deep reinforcement learning (DRL) has successfully addressed many complex control problems. However, the neural networks representing policies or values remain opaque, undermining trust in high-stakes applications. While concept-based methods have shown promise in deciphering internal representation…

• How Should World Models Be Evaluated for Embodied Decision-Making? A Decision-Making-Centric Position

  Yang Yu, Shiyuan Zhang, Yifei Sheng, Haoxiang Ren, Haoxin Lin · 30 juin 2026

  World models have become a central abstraction in modern AI. The term now refers to several different objects: action-conditioned environment models, latent imagination models, future-video predictors, interactive neural simulators, latent predictive representations, and synthetic-data engines. Eval…

• FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization

  Junyan Tan, Yifan Li, Minghao Wang, Zihan Chen, Haoyu Zhang · 30 juin 2026

  Venture capital (VC) investment decisions face distinct challenges, such as multi-source heterogeneous data, non-stationary time series, and the demand for explainable predictions in high-stakes, low-data settings. To overcome these issues, we introduce \textbf{FinInvest-GTCN}, a Graph-Temporal-Caus…

• Gradient boosting with vector-valued leafs

  David Cortes · 30 juin 2026

  Gradient boosting in the form of decision tree ensembles has successfully been applied to a variety of problems using simple objective functions based on log-likelihoods of a single variable. The concept extends naturally to objective functions operating on vectors - for example, multinomial logisti…

• What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs

  Nhi Nguyen, Shauli Ravfogel, Rajesh Ranganath · 30 juin 2026

  Large language models (LLMs) are increasingly deployed in high-stakes domains, where free-text explanations such as chain-of-thought and post-hoc rationales are used to justify model outputs. Yet it remains unclear whether these explanations are sufficient, i.e., if they contain enough information t…