Explainable Artificial Intelligence (XAI)

Derniers papiers

• The Representational Limit of Scalar Interactions: An Interventional Decomposition

  Potito Aghilar, Sabino Roccotelli, Stanislao Fidanza, Vito Walter Anelli, Sebastiano Stramaglia, Tommaso Di Noia · 19 juin 2026

  Signed pairwise interaction scores fundamentally conflate uniqueness (U), redundancy (R), and synergy (S). We prove this on a minimal 3-way XOR structural causal model: faithful indices such as Shapley-Taylor return zero per pair, whereas projective indices such as Shapley Interaction spread the thi…

• Multimodal Concept Bottleneck Models

  Tongqing Shi, Ge Yan, Tuomas Oikarinen, Tsui-Wei Weng · 19 juin 2026

  Concept Bottleneck Models (CBMs) enhance the interpretability of deep learning networks by aligning the features extracted from images with natural concepts. However, existing CBMs are constrained in their ability to generalize beyond a fixed set of predefined classes and the risk of non-concept inf…

• Concept Flow Models: Anchoring Concept-Based Reasoning with Hierarchical Bottlenecks

  Ya Wang, Adrian Paschke · 19 juin 2026

  Concept Bottleneck Models (CBMs) enhance interpretability by projecting learned features into a human-understandable concept space. Recent approaches leverage vision-language models to generate concept embeddings, reducing the need for manual concept annotations. However, these models suffer from a …

• DeepSWIP: Quotient-WMC Counterfactuals for Neural Probabilistic Logic Programs

  Saimun Habib, Vaishak Belle, Fengxiang He · 19 juin 2026

  Neurosymbolic systems such as DeepProbLog combine neural perception with probabilistic logic, but standard inference is associational. Counterfactual reasoning additionally requires a causal semantics for interventions and evidence. We introduce DeepSWIP, a single-world counterfactual semantics for …

• Neural Additive and Basis Models with Feature Selection and Interactions

  Yasutoshi Kishimoto, Kota Yamanishi, Takuya Matsuda, Shinichi Shirakawa · 19 juin 2026

  Deep neural networks (DNNs) exhibit attractive performance in various fields but often suffer from low interpretability. The neural additive model (NAM) and its variant called the neural basis model (NBM) use neural networks (NNs) as nonlinear shape functions in generalized additive models (GAMs). B…

• ADaPT: Token-Level Decoupling for Efficient Large Reasoning Models

  Tingyun Li, Zishang Jiang, Jinyi Han, Xinyi Wang, Sihang Jiang, Han Xia, Zhaoqian Dai, Shuguang Ma, Fei Yu, Jiaqing Liang, Yanghua Xiao · 19 juin 2026

  Large reasoning models rely on long chain-of-thought to achieve strong performance, but applying such reasoning uniformly incurs high computational cost. Existing efficiency-oriented methods attempt to shorten or mix reasoning strategies, yet often degrade reasoning capability. We identify the root …

• Beyond Uniform Forgetting: A Study of Sequential Direct Preference Optimization Across Preference Settings

  Pranav Bhandari, Nicolas Fay, Amitava Datta, Usman Naseem, Mehwish Nasim · 19 juin 2026

  Aligning language models with human preferences often requires optimising multiple behavioural objectives. A practical approach is to apply these objectives sequentially using preference optimisation methods such as Direct Preference Optimisation (DPO), but it remains unclear whether later training …

• GLARE: A Natural Language Interface for Querying Global Explanations

  Bhavan Vasu, Rajesh Mangannavar · 19 juin 2026

  While global explanations are crucial for understanding vision models across datasets, classes, and decision contexts, their complex and monolithic nature often hinders practical exploration. Because users typically seek targeted answers to specific questions rather than static artifacts, we present…

• Interpreting Neural Combinatorial Optimization via Evolving Programmatic Bottlenecks

  Haocheng Duan, Yuxin Guo, Jieyi Bi, Anqi Xie, Sirui Li, Yining Ma, Cathy Wu · 19 juin 2026

  Neural Combinatorial Optimization (NCO) achieves strong performance, yet its black-box nature remains a key roadblock to deployment and scientific diagnosis. Standard interpretability tools, such as Concept Bottleneck Models (CBMs), are ill-equipped for NCO, whose decisions are dynamic, state-depend…

• Interpretable and Verifiable Hardware Generation with LLM-Driven Stepwise Refinement

  You Li, Samuel Mandell, David Z. Pan · 19 juin 2026

  Large language models (LLMs) have achieved remarkable success in software development. However, they are susceptible to hallucinations, meaning that they can introduce subtle semantic and logical errors. Due to the high stakes in chip design and manufacturing, hardware engineers are still reluctant …

• Critical Percolation as a Synthetic Data Model for Interpretability

  Aryeh Brill, Tom Ingebretsen Carlson · 19 juin 2026

  Neural networks learn features that reflect the hierarchical, multi-scale structure of natural data. Synthetic datasets used to evaluate interpretability methods typically lack this structure, limiting their value as realistic toy models. To close this gap, we introduce a family of synthetic dataset…

• JustDiag!: A Diagnostic Justification Engine for Accountable Root Cause Analysis

  Tingzhu Bi, Xinrui Jiang, Xun Zhang, Pengcheng Su, Congjie He, Jinglin Li, Ping Wang, Meng Ma · 19 juin 2026

  Large language models can produce fluent root cause analyses, but fluent final answers alone are insufficient evidence for accountability in high-stakes operations. In real incident response, engineers need to know what evidence supported a diagnosis, which alternatives were considered, where contra…

• Manifold Bandits: Bayesian Curriculum Learning over the Latent Geometry of Large Language Models

  Darrien McKenzie, Nicklas Hansen, Xiaolong Wang · 19 juin 2026

  Reinforcement learning (RL) is a central approach for improving reasoning capabilities in large language models (LLMs), where training efficiency depends critically on how problems are sampled during optimization. Existing adaptive curriculum learning methods typically prioritize prompts of intermed…

• Comparing Linear Probes with Mahalanobis Cosine Similarity

  Zhuofan Josh Ying, Peter Hase, Nikolaus Kriegeskorte · 19 juin 2026

  Linear probes are widely used in interpretability research and often compared by cosine similarity. The Mahalanobis cosine similarity (MCS) between two directions, which reweights the inner product by test data covariance, is a natural task-aware refinement. Ying et al. (2026) report that a probe's …

• SAGE: Retain-Aware Post-Hoc Sanitization of Final Unlearning Vector

  Jingyuan Zhang, Yucheng Bai, Peixi Wen, Zhehao Huang, Zhengbao He, Hanling Tian, Xinwen Cheng, Haiyin Ran, Xiaolin Huang · 18 juin 2026

  Large Language Model (LLM) unlearning aims to remove undesirable knowledge or behaviors while preserving retained capabilities. Current unlearning methods all involve a trade-off between unlearning and retention. We have found that the retention activation bias can also be used to quantify the damag…

• Kernel of Partition Paths: A Unified Representation for Tree Ensembles

  Nicolas Mahler · 18 juin 2026

  A recent line of work has reframed individual decision trees as linear models on engineered features associated with their splits, opening routes for oracle inequalities and feature-importance reinterpretation, but leaving open the question of what unified geometric object a forest induces when one …

• P$^2$CE: Model-Agnostic Plausible Pareto-Optimal Counterfactual Explanations

  Arthur Hendricks Mendes de Oliveira, Giovani Valdrighi, Marcos Medeiros Raimundo · 18 juin 2026

  The increasing use of machine learning algorithms in social applications has raised concerns about fairness and transparency, leading to the development of counterfactual explanations. These explanations supports individuals to understand and potentially alter unfavorable decisions in areas such as …

• From Mechanistic to Compositional Interpretability

  Ward Gauderis, Thomas Dooms, Steven T. Homer, Kola Ayonrinde, Geraint A. Wiggins · 18 juin 2026

  Mechanistic interpretability aims to explain neural model behaviour by reverse-engineering learned computational structure into human-understandable components. Without a formal framework, however, mechanistic explanations cannot be objectively verified, compared, or composed. We introduce compositi…

• Effects of sparsity and superposition on loss in simple autoencoders

  Mriganka Basu Roy Chowdhury, Eric McLaughlin Weiner · 18 juin 2026

  One of the major difficulties in the mechanistic interpretability of neural networks is the occurrence of polysemanticity, which suggests that each neuron is typically responsible for multiple different tasks, impeding a clean interpretation of their function. The seminal paper of Elhage et al. (202…

• Target-confidence Recourse Using tSeTlin machines: TRUST

  K. Darshana Abeyrathna, Sara El Mekkaoui, Nils Enric Canut Taugb{\o}l, Anuja Vats · 18 juin 2026

  Counterfactual explanations are widely used to provide algorithmic recourse in high-stakes decision-making systems. Most existing methods seek the smallest change to an input that flips a model's decision. However, decision-makers often rely not only on predicted labels but also on confidence thresh…

• Strategic Feature Selection

  Jivat Neet Kaur, Pratik Patil, Divya Shanmugam, Emma Pierson, Michael I. Jordan, Nika Haghtalab, Meena Jagadeesan, Ahmed Alaa, Serena Wang · 18 juin 2026

  When algorithmic predictors inform resource allocation in high-stakes domains such as healthcare, these predictors must account for strategic manipulation of input features. The typical solution is to redesign the predictor itself to explicitly account for strategic interactions. In practice, howeve…

• Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systems

  Hehai Lin, Qi Yang, Chengwei Qin · 18 juin 2026

  Large Language Model (LLM)-based automatic Multi-Agent Systems (MAS) generation has become a crucial frontier for tackling complex tasks. However, existing methods face a dilemma between model capability and experience retention. Inference-time MAS leverages frozen frontier LLMs but repeats identica…

• From Sparse Features to Trustworthy Proxies: Certifying SAE-Based Interpretability

  Dibyanayan Bandyopadhyay, Asif Ekbal · 18 juin 2026

  Sparse autoencoders (SAEs) are increasingly used to extract interpretable features from language models (LMs), yet a central question remains: when can an SAE-based explanation be treated as a faithful view of an underlying frozen LM We study this through a post-hoc generalization framework that cer…

• Explaining Attention with Program Synthesis

  Amiri Hayes, Belinda Li, Jacob Andreas · 18 juin 2026

  A longstanding goal of research on interpretable deep learning is to replace opaque neural computations with human-meaningful symbolic descriptions. In this paper, we propose an approach for approximating the behavior of components of deep networks with executable programs. We focus on attention hea…

• Self-CTRL: Self-Consistency Training with Reinforcement Learning

  Itamar Pres, Laura Ruis, Melat Ghebreselassie, Belinda Z. Li, Jacob Andreas · 18 juin 2026

  Language models (LMs) that faithfully describe their own behavior can more easily be audited, understood, and trusted by users. This paper describes Self-Consistency Training with Reinforcement Learning (Self-CTRL), a method that optimizes for consistency between a LM's self-explanations and behavio…