Explainable Artificial Intelligence (XAI)

Derniers papiers

• Single-Position Intervention Fails: Distributed Output Templates Drive In-Context Learning

  Bryan Cheng, Jasper Zhang · 7 mai 2026

  Understanding how large language models encode task identity from few-shot demonstrations is a central open problem in mechanistic interpretability. Prior work uses linear probing to localize task representations, reporting high classification accuracy at specific layers. We reveal a striking dissoc…

• NoisyCausal: A Benchmark for Evaluating Causal Reasoning Under Structured Noise

  Zhi Xu, Yun Fu · 7 mai 2026

  Causal reasoning in natural language requires identifying relevant variables, understanding their interactions, and reasoning about effects and interventions, often under noisy or ambiguous conditions. While large language models (LLMs) exhibit strong general reasoning abilities, they struggle to di…

• Toward Human-AI Complementarity Across Diverse Tasks

  Yuzheng Xu, Annya Dahmani, Matthew D. Blanchard, Niclas Dern, Edy Nastase, Francesca Bianco, Maja Pavlovic, Sukanya Krishna, Eric Modesitt, Miranda Anna Christ, Arth Singh, Gaia Molinaro, Sikata Bela Sengupta, Jaji Pamarthi, Arjun Menon, Rishub Jain · 7 mai 2026

  Human-AI complementarity, the idea that combining human and AI judgments can outperform either alone, offers a promising pathway toward robust oversight of advanced AI systems. However, whether human-AI complementarity can be achieved on realistic tasks remains an open question. We investigate this …

• Quadrature-TreeSHAP: Depth-Independent TreeSHAP and Shapley Interactions

  Ron Wettenstein, Rory Mitchell, Peng Yu · 7 mai 2026

  Shapley values are a standard tool for explaining predictions of tree ensembles, with Path-Dependent SHAP being the most widely used variant. Despite substantial progress, existing methods still exhibit trade-offs between depth-dependent runtime, numerical stability, and support for higher-order int…

• Evaluation Cards for XAI Metrics

  Rokas Gipi\v{s}kis, Olga Kurasova · 7 mai 2026

  The evaluation of explainable AI (XAI) methods is affected by a lack of standardization. Metrics are inconsistently defined, incompletely reported, and rarely validated against common baselines. In this paper, we identify transparency of evaluation reporting as a central, under-addressed problem. We…

• Gyan: An Explainable Neuro-Symbolic Language Model

  Venkat Srinivasan, Vishaal Jatav, Anushka Chandrababu, Geetika Sharma · 7 mai 2026

  Transformer based pre-trained large language models have become ubiquitous. There is increasing evidence to suggest that even with large scale pre-training, these models do not capture complete compositional context and certainly not, the full human analogous context. Besides, by the very nature of …

• Think-Aloud Reshapes Automated Cognitive Model Discovery Beyond Behavior

  Hanbo Xie, Akshay K. Jagadish, Lan Pan, Robert C. Wilson · 7 mai 2026

  Computational cognitive models discovered using large language models have so far relied solely on behavioral data. However, it is well-known that models produced from the behavioral trajectory alone are typically under-determined. In this work, we explore the use of Think Aloud traces as an additio…

• Sequential Strategic Classification with Multi-Stage Selective Classifiers

  Ziyuan Huang, Lina Alkarmi, Mingyan Liu · 7 mai 2026

  Strategic classification studies the problem where self-interested individuals or agents manipulate their response to obtain favorable decision outcomes made by classifiers, typically turning to dishonest actions when they are less costly than genuine efforts. Prior works have demonstrated a fundame…

• Decision Evidence Maturity Model for Agentic AI: A Property-Level Method Specification

  Oleg Solozobov · 7 mai 2026

  Agentic AI systems produce decision evidence at scale through execution telemetry, but property-level reconstruction often fails when an external party asks a specific governance question about a specific decision: the assembled evidence is insufficient to answer it. We name this pattern the contain…

• LLM-Powered AI Agent Systems and Their Applications in Industry

  Guannan Liang, Qianqian Tong · 6 mai 2026

  The emergence of Large Language Models (LLMs) has reshaped agent systems. Unlike traditional rule-based agents with limited task scope, LLM-powered agents offer greater flexibility, cross-domain reasoning, and natural language interaction. Moreover, with the integration of multi-modal LLMs, current …

• Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution

  Ruta Binkyte, Ivaxi Sheth, Zhijing Jin, Mohammad Havaei, Bernhard Sch\"olkopf, Mario Fritz · 6 mai 2026

  As artificial intelligence (AI), including machine learning (ML) models and foundation models (FMs), is increasingly deployed in high-stakes domains, ensuring their trustworthiness has become a central challenge. However, the core trustworthy AI objectives, such as fairness, robustness, privacy, and…

• Measuring AI Reasoning: A Guide for Researchers

  Munachiso Samuel Nwadike, Zangir Iklassov, Kareem Ali, Rifo Genadi, Kentaro Inui · 6 mai 2026

  In this paper, we offer a guide for researchers on evaluating reasoning in language models, building the case that reasoning should be assessed through evidence of adaptive, multi-step search rather than final-answer accuracy alone. Under an evaluation-oriented definition, reasoning requires selecti…

• Hey, That's My Data! Token-Only Dataset Inference in Large Language Models

  Chen Xiong, Zihao Wang, Rui Zhu, Tsung-Yi Ho, Pin-Yu Chen, Jingwei Xiong, Haixu Tang · 6 mai 2026

  Large Language Models (LLMs) rely on massive training datasets, often including proprietary data, which raises concerns about unauthorized usage and copyright infringement. Existing dataset inference methods typically require access to log probabilities or other internal signals, but many modern LLM…

• When Stress Becomes Signal: Detecting Antifragility-Compatible Regimes in Multi-Agent LLM Systems

  Jose Manuel de la Chica, Juan Manuel Vera, Jairo Rodr\'iguez · 6 mai 2026

  Multi-agent LLM systems are increasingly used to solve complex tasks through decomposition, debate, specialization, and ensemble reasoning. However, these systems are usually evaluated in terms of robustness: whether performance is preserved under perturbation. This paper studies a different questio…

• Coherent Hierarchical Multi-Label Learning to Defer for Medical Imaging

  Joshua Strong, Pramit Saha, Emma Sun, Helen Higham, Alison Noble · 6 mai 2026

  Learning to Defer (L2D) enables a model to predict autonomously or defer to an expert, but prior work largely assumes flat label spaces. We study the first L2D setting with hierarchical multi-label decisions, motivated by medical-imaging workflows in which findings are organised by clinical taxonomi…

• Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agents

  Bowen Ye, Rang Li, Qibin Yang, Yuanxin Liu, Linli Yao, Hanglong Lv, Zhihui Xie, Chenxin An, Lei Li, Lingpeng Kong, Qi Liu, Zhifang Sui, Tong Yang · 6 mai 2026

  Large language models are increasingly deployed as autonomous agents for multi-step workflows in real-world software environments. However, existing agent benchmarks are limited by trajectory-opaque grading, underspecified safety and robustness evaluation, and narrow coverage of modalities and inter…

• Toward a Science of Intent: Closure Gaps and Delegation Envelopes for Open-World AI Agents

  Maximiliano Armesto, Christophe Kolb · 6 mai 2026

  Recent work has framed intelligence in verifiable tasks as reducing time-to-solution through learned structure and test-time search, while systems work has explored learned runtimes in which computation, memory and I/O migrate into model state. These perspectives do not explain why capable models re…

• When Should a Language Model Trust Itself? Same-Model Self-Verification as a Conditional Confidence Signal

  Aditya Ajay Phalod · 6 mai 2026

  Same-model self-verification, prompting a model to audit its own predicted answer, is a plausible confidence signal for selective prediction, but its practical value remains unclear once strong likelihood-based baselines are taken seriously. We evaluate self-verification against two such baselines, …

• Less Interaction But More Explanation: A Communication Perspective on Agentic AI Interfaces

  Eunchae Jang, S. Shyam Sundar · 6 mai 2026

  AI systems have long been expected to interact with users, answering questions, generating content, and continuing (social) conversations. Agentic AI, however, breaks from this expectation, as its primary objective is workflow execution on behalf of the users. If a system becomes more agentic, do us…

• ValueBlindBench: Agreement-Gated Stress Testing of LLM-Judged Investment Rationales Before Returns Are Observable

  Sidi Chang, Peiying Zhu, Yuxiao Chen · 6 mai 2026

  LLM-based financial agents increasingly produce investment rationales before the outcomes needed to evaluate them are observable. This creates a delayed-ground-truth evaluation problem: realized returns remain the eventual arbiter of investment quality, but they arrive too late and are too noisy to …

• Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction

  Li Puyin, Jiyuan Tan, Ahmad Jabbar, Thomas Icard, Atticus Geiger · 6 mai 2026

  We present a method for diagnosing interpretation in neural networks by identifying an input subspace where a proposed interpretation is highly faithful. Our method is particularly useful for causal-abstraction-style interpretability, where a high-level causal hypothesis is evaluated by interchange …

• Automated Interpretability and Feature Discovery in Language Models with Agents

  Arnau Marin-Llobet, Javier Ferrando · 6 mai 2026

  We introduce an autonomous multiagent framework for mechanistic interpretability that automates both explaining and finding internal features in large language models. The system runs two coupled loops: (1) explanation refinement, where an agent proposes competing hypotheses and iteratively tests th…

• First-Order Efficiency for Probabilistic Value Estimation via A Statistical Viewpoint

  Ziqi Liu, Kiljae Lee, Yuan Zhang, Weijing Tang · 6 mai 2026

  Probabilistic values, including Shapley values and semivalues, provide a model-agnostic framework to attribute the behavior of a black-box model to data points or features, with a wide range of applications including explainable artificial intelligence and data valuation. However, their exact comput…

• Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation

  Francesco Sovrano, Gabriele Dominici, Marc Langheinrich · 6 mai 2026

  A key goal of explainable AI (XAI) is to express the decision logic of large language models (LLMs) in symbolic form and link it to internal mechanisms. Global rule-extraction methods typically learn symbolic surrogates without grounding rules in model circuitry, while mechanistic interpretability c…

• Multiple Choice Questions: Reasoning Makes Large Language Models (LLMs) More Self-Confident, Especially When They are Wrong

  Tairan Fu, Javier Conde, Gonzalo Mart\'inez, Mar\'ia Grandury, Pedro Reviriego · 6 mai 2026

  Multiple Choice Question (MCQ) tests are among the most used methods for evaluating large language models (LLMs). Besides checking the correctness of the selected answer, evaluations often consider the model's confidence through the probability assigned to its response. In this work, we investigate …