Explainable Artificial Intelligence (XAI)

Derniers papiers

• Towards Identifiable Latent Additive Noise Models

  Yuhang Liu, Zhen Zhang, Dong Gong, Erdun Gao, Biwei Huang, Mingming Gong, Anton van den Hengel, Kun Zhang, Javen Qinfeng Shi · 30 janvier 2026

  Causal representation learning (CRL) offers the promise of uncovering the underlying causal model by which observed data was generated, but the practical applicability of existing methods remains limited by the strong assumptions required for identifiability and by challenges in applying them to rea…

• Sycophantic Anchors: Localizing and Quantifying User Agreement in Reasoning Models

  Jacek Duszenko · 30 janvier 2026

  Reasoning models frequently agree with incorrect user suggestions -- a behavior known as sycophancy. However, it is unclear where in the reasoning trace this agreement originates and how strong the commitment is. To localize and quantify this behavior, we introduce \emph{sycophantic anchors} -- sent…

• TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learning

  Shicheng Fan, Kun Zhang, Lu Cheng · 30 janvier 2026

  Temporal causal representation learning methods assume that causal mechanisms switch instantaneously between discrete domains, yet real-world systems often exhibit continuous mechanism transitions. For example, a vehicle's dynamics evolve gradually through a turning maneuver, and human gait shifts s…

• Beyond Forgetting: Machine Unlearning Elicits Controllable Side Behaviors and Capabilities

  Tien Dang, The-Hai Nguyen, Dinh Mai Phuong, Nguyen Minh Phuong, Hoang Thanh-Tung, Le-Minh Nguyen, Naoya Inoue · 30 janvier 2026

  We consider representation misdirection (RM), a class of LLM unlearning methods that achieves forgetting by manipulating the forget-representations, that is, latent representations of forget samples. Despite being important, the roles of target vectors used in RM, however, remain underexplored. Here…

• ECSEL: Explainable Classification via Signomial Equation Learning

  Adia Lumadjeng, Ilker Birbil, Erman Acar · 30 janvier 2026

  We introduce ECSEL, an explainable classification method that learns formal expressions in the form of signomial equations, motivated by the observation that many symbolic regression benchmarks admit compact signomial structure. ECSEL directly constructs a structural, closed-form expression that ser…

• World of Workflows: a Benchmark for Bringing World Models to Enterprise Systems

  Lakshya Gupta (Skyfall AI), Litao Li (Skyfall AI), Yizhe Liu (Skyfall AI), Sriram Ganapathi Subramanian (Skyfall AI), Kaheer Suleman (Skyfall AI), Zichen Zhang (Skyfall AI), Haoye Lu (Skyfall AI), Sumit Pasupalak (Skyfall AI) · 30 janvier 2026

  Frontier large language models (LLMs) excel as autonomous agents in many domains, yet they remain untested in complex enterprise systems where hidden workflows create cascading effects across interconnected databases. Existing enterprise benchmarks evaluate surface-level agentic task completion simi…

• A Unified Theory of Sparse Dictionary Learning in Mechanistic Interpretability: Piecewise Biconvexity and Spurious Minima

  Yiming Tang, Harshvardhan Saini, Zhaoqian Yao, Zheng Lin, Yizhen Liao, Qianxiao Li, Mengnan Du, Dianbo Liu · 30 janvier 2026

  As AI models achieve remarkable capabilities across diverse domains, understanding what representations they learn and how they encode concepts has become increasingly important for both scientific progress and trustworthy deployment. Recent works in mechanistic interpretability have widely reported…

• Causal Discovery for Explainable AI: A Dual-Encoding Approach

  Henry Salgado, Meagan R. Kendall, Martine Ceberio · 30 janvier 2026

  Understanding causal relationships among features is fundamental for explaining machine learning model decisions. However, traditional causal discovery methods face challenges with categorical variables due to numerical instability in conditional independence testing. We propose a dual-encoding caus…

• SymbXRL: Symbolic Explainable Deep Reinforcement Learning for Mobile Networks

  Abhishek Duttagupta, MohammadErfan Jabbari, Claudio Fiandrino, Marco Fiore, Joerg Widmer · 30 janvier 2026

  The operation of future 6th-generation (6G) mobile networks will increasingly rely on the ability of deep reinforcement learning (DRL) to optimize network decisions in real-time. DRL yields demonstrated efficacy in various resource allocation problems, such as joint decisions on user scheduling and …

• Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units

  Jianhui Chen, Yuzhang Luo, Liangming Pan · 30 janvier 2026

  While Mechanistic Interpretability has identified interpretable circuits in LLMs, their causal origins in training data remain elusive. We introduce Mechanistic Data Attribution (MDA), a scalable framework that employs Influence Functions to trace interpretable units back to specific training sample…

• Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models

  Konstantinos P. Panousis, Diego Marcos · 30 janvier 2026

  The widespread adoption of Vision-Language Models (VLMs) across fields has amplified concerns about model interpretability. Distressingly, these models are often treated as black-boxes, with limited or non-existent investigation of their decision making process. Despite numerous post- and ante-hoc i…

• Enhancing Language Models for Robust Greenwashing Detection

  Neil Heinrich Braun, Keane Ong, Rui Mao, Erik Cambria, Gianmarco Mengaldo · 30 janvier 2026

  Sustainability reports are critical for ESG assessment, yet greenwashing and vague claims often undermine their reliability. Existing NLP models lack robustness to these practices, typically relying on surface-level patterns that generalize poorly. We propose a parameter-efficient framework that str…

• Modeling Endogenous Logic: Causal Neuro-Symbolic Reasoning Model for Explainable Multi-Behavior Recommendation

  Yuzhe Chen, Jie Cao, Youquan Wang, Haicheng Tao, Darko B. Vukovic, Jia Wu · 30 janvier 2026

  Existing multi-behavior recommendations tend to prioritize performance at the expense of explainability, while current explainable methods suffer from limited generalizability due to their reliance on external information. Neuro-Symbolic integration offers a promising avenue for explainability by co…

• AgenticSimLaw: A Juvenile Courtroom Multi-Agent Debate Simulation for Explainable High-Stakes Tabular Decision Making

  Jon Chun, Kathrine Elkins, Yong Suk Lee · 30 janvier 2026

  We introduce AgenticSimLaw, a role-structured, multi-agent debate framework that provides transparent and controllable test-time reasoning for high-stakes tabular decision-making tasks. Unlike black-box approaches, our courtroom-style orchestration explicitly defines agent roles (prosecutor, defense…

• Reward Models Inherit Value Biases from Pretraining

  Brian Christian, Jessica A. F. Thompson, Elle Michelle Yang, Vincent Adam, Hannah Rose Kirk, Christopher Summerfield, Tsvetomira Dumbalska · 29 janvier 2026

  Reward models (RMs) are central to aligning large language models (LLMs) with human values but have received less attention than pre-trained and post-trained LLMs themselves. Because RMs are initialized from LLMs, they inherit representations that shape their behavior, but the nature and extent of t…

• Structural Compositional Function Networks: Interpretable Functional Compositions for Tabular Discovery

  Fang Li · 29 janvier 2026

  Despite the ubiquity of tabular data in high-stakes domains, traditional deep learning architectures often struggle to match the performance of gradient-boosted decision trees while maintaining scientific interpretability. Standard neural networks typically treat features as independent entities, fa…

• GRACE: A Language Model Framework for Explainable Inverse Reinforcement Learning

  Silvia Sapora, Devon Hjelm, Alexander Toshev, Omar Attia, Bogdan Mazoure · 29 janvier 2026

  Inverse Reinforcement Learning aims to recover reward models from expert demonstrations, but traditional methods yield black-box models that are difficult to interpret and debug. In this work, we introduce GRACE (Generating Rewards As CodE), a method for using Large Language Models within an evoluti…

• Optimal Transport Group Counterfactual Explanations

  Enrique Valero-Leal, Bernd Bischl, Pedro Larra\~naga, Concha Bielza, Giuseppe Casalicchio · 29 janvier 2026

  Group counterfactual explanations find a set of counterfactual instances to explain a group of input instances contrastively. However, existing methods either (i) optimize counterfactuals only for a fixed group and do not generalize to new group members, (ii) strictly rely on strong model assumption…

• GNN Explanations that do not Explain and How to find Them

  Steve Azzolin, Stefano Teso, Bruno Lepri, Andrea Passerini, Sagar Malhotra · 29 janvier 2026

  Explanations provided by Self-explainable Graph Neural Networks (SE-GNNs) are fundamental for understanding the model's inner workings and for identifying potential misuse of sensitive attributes. Although recent works have highlighted that these explanations can be suboptimal and potentially mislea…

• Fair Recourse for All: Ensuring Individual and Group Fairness in Counterfactual Explanations

  Fatima Ezzeddine, Obaida Ammar, Silvia Giordano, Omran Ayoub · 29 janvier 2026

  Explainable Artificial Intelligence (XAI) is becoming increasingly essential for enhancing the transparency of machine learning (ML) models. Among the various XAI techniques, counterfactual explanations (CFs) hold a pivotal role due to their ability to illustrate how changes in input features can al…

• Higher-Order Feature Attribution: Bridging Statistics, Explainable AI, and Topological Signal Processing

  Kurt Butler, Guanchao Feng, Petar Djuric · 29 janvier 2026

  Feature attributions are post-training analysis methods that assess how various input features of a machine learning model contribute to an output prediction. Their interpretation is straightforward when features act independently, but it becomes less clear when the predictive model involves interac…

• Demystifying Prediction Powered Inference

  Yilin Song, Dan M. Kluger, Harsh Parikh, Tian Gu · 29 janvier 2026

  Machine learning predictions are increasingly used to supplement incomplete or costly-to-measure outcomes in fields such as biomedical research, environmental science, and social science. However, treating predictions as ground truth introduces bias while ignoring them wastes valuable information. P…

• Memory Retrieval in Transformers: Insights from The Encoding Specificity Principle

  Viet Hung Dinh, Ming Ding, Youyang Qu, Kanchana Thilakarathna · 29 janvier 2026

  While explainable artificial intelligence (XAI) for large language models (LLMs) remains an evolving field with many unresolved questions, increasing regulatory pressures have spurred interest in its role in ensuring transparency, accountability, and privacy-preserving machine unlearning. Despite re…

• Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs

  Yuhang Liu, Erdun Gao, Dong Gong, Anton van den Hengel, Javen Qinfeng Shi · 29 janvier 2026

  Developing human understandable interpretation of large language models (LLMs) becomes increasingly critical for their deployment in essential domains. Mechanistic interpretability seeks to mitigate the issues through extracts human-interpretable process and concepts from LLMs' activations. Sparse a…

• The Grammar of Transformers: A Systematic Review of Interpretability Research on Syntactic Knowledge in Language Models

  Nora Graichen, Iria de-Dios-Flores, Gemma Boleda · 29 janvier 2026

  We present a systematic review of 337 articles evaluating the syntactic abilities of Transformer-based language models, reporting on 1,015 model results from a range of syntactic phenomena and interpretability methods. Our analysis shows that the state of the art presents a healthy variety of method…