Explainable Artificial Intelligence (XAI)

Derniers papiers

• SpheriCity: Designing Trustworthy Conversational AI for Sustainability Decision Support

  Ahmed Qayyum, Madison Werner, Kathryn Youngblood, Jenna R. Jambeck, Tahiya Chowdhury · 15 juin 2026

  We present SpheriCity, an expert-grounded conversational prototype designed to support trustworthy knowledge sensemaking from sustainability reports. City-level circularity assessment reports contain rich information about materials, infrastructure, and policy interventions, yet their length and het…

• Generative AI for Managerial Decision-Making under Ambiguity and Sycophancy

  Sule Ozturk Birim, Fabrizio Marozzo, Yigit Kazancoglu · 15 juin 2026

  Generative artificial intelligence (GenAI) is increasingly being integrated into complex business workflows, fundamentally shifting the boundaries of managerial decision-making. However, the reliability of its strategic advice in ambiguous business contexts remains a critical knowledge gap. To addre…

• The Perceived Fragility of Explanations in Audio Models: Manipulation of Attribution with Unchanged Predictions

  Piotr Kit{\l}owski, Dominik Wi\k{a}cek, Mateusz Modrzejewski · 15 juin 2026

  This paper investigates the fragility of post-hoc explanation methods in audio deepfake detection. While previous work on explanation manipulation focused on images using standard $L_p$ metrics, we introduce a psychoacoustic framework that optimizes inaudible perturbations to decouple model attribut…

• Capability Minimization as a Safety Primitive: Risk-Aware Causal Gating for Least-Privilege LLM Agents

  Laxmipriya Ganesh Iyer, Rahul Suresh Babu · 15 juin 2026

  Modern decision systems increasingly rely on learned components whose outputs may be confident yet wrong, exposing downstream actions to costly errors. We introduce Risk-Aware Causal Gating (RACG), a framework that decides whether to act on, defer, or abstain from a model's prediction by combining c…

• Distributional Biases in Post-Training: A Markovian Analysis of Reasoning Trajectories

  Dake Bu, Wei Huang, Andi Han, Atsushi Nitanda, Bo Xue, Qingfu Zhang, Hau-San Wong, Taiji Suzuki · 15 juin 2026

  Foundation models exhibit broad knowledge but limited task-specific reasoning, motivating post-training strategies such as RL with verifiable rewards (RLVR) and test-time scaling (TTS). While recent work highlights the role of exploration in improving pass@K, empirical evidence points to a paradox: …

• Ensembling Sparse Autoencoders

  Soham Gadgil, Chris Lin, Su-In Lee · 15 juin 2026

  Sparse autoencoders (SAEs) are used to decompose neural network activations into human-interpretable features. Typically, features learned by a single SAE are used for downstream applications. However, it has recently been shown that a single SAE captures only a limited subset of features that can b…

• Learning High Coverage Discriminative Parsimonious Rulesets

  Mariamma Antony, Raman Sankaran, Chiranjib Bhattacharyya, Uma Satya Ranjan · 15 juin 2026

  Learning systems based on IF-THEN rule representations readily offer interpretability, making them a crucial focus in contemporary AI research. A key objective for such rule sets is to achieve both high discriminative power and interpretability. While existing state-of-the-art algorithms implicitly …

• Listening with Attention: Entropy-Guided Explainability for Transformer-Based Audio Models

  Ravi Ranjan, Utkarsh Grover, Xiaomin Lin, Agoritsa Polyzou · 15 juin 2026

  Transformer-based automatic speech recognition (ASR) models such as Whisper are highly accurate, but their predictions remain difficult to interpret. Existing explainable AI (XAI) methods often lack faithfulness and precise temporal grounding. We propose Listening with Entropy-guided Attention for F…

• EM-NeSy: Expectation Maximization for Neurosymbolic Learning

  Annegret Seibt, Luc De Raedt, Giuseppe Marra · 15 juin 2026

  Neurosymbolic (NeSy) models integrate neural networks and symbolic reasoning for robust and interpretable AI. State-of-the-art NeSy models require that the symbolic component is expressed in a differentiable way, often complicating the use of approximate inference. We propose EM-NeSy which casts pro…

• Abstracting Cross-Domain Action Sequences into Interpretable Workflows

  Gaurav Verma, Scott Counts · 15 juin 2026

  Sequential or time-stamped interaction logs provide objective records of digital application usage, yet their granularity and noise often obscure meaningful insights into people's work. Such insights are essential for improving digital products in ways grounded in real-world user interactions. Prior…

• Decompose Sparsely Where You Should, Absorb Densely Where You Should No

  Ruixuan Deng, Zehao Jin, Zekun Wang, Zihan Dong · 15 juin 2026

  Sparse autoencoders (SAEs) are typically trained to reconstruct the \textbf{entire} residual stream through a sparse dictionary, implicitly assuming that all activation content is amenable to sparse, monosemantic decomposition. We question this assumption and hypothesize that activations contain a l…

• ReSET: Accurate Latency-Critical NVFP4 Reasoning via Step-Aware Temperature Scaling

  Sihwa Lee, Janghwan Lee, Donghoon Yoo, Jae Gon Kim, Hanyul Ryu, Soojung Ryu, Jungwook Choi · 12 juin 2026

  Large reasoning models (LRMs) improve complex problem-solving by generating long intermediate reasoning traces, but this substantially increases inference costs. NVFP4 inference offers a promising approach to reduce both computational and memory costs through hardware-supported low-precision executi…

• Decoding the Multimodal Maze: A Systematic Review on the Adoption of Explainability in Multimodal Attention-based Models

  Md Raisul Kibria, S\'ebastien Lafond, Janan Arslan · 12 juin 2026

  Multimodal learning has witnessed remarkable advancements in recent years, particularly with the integration of attention-based models, leading to significant performance gains across a variety of tasks. Parallel to this progress, the demand for explainable artificial intelligence (XAI) has spurred …

• The Query Channel: Information-Theoretic Limits of Masking-Based Explanations

  Erciyes Karakaya, Ozgur Ercetin · 12 juin 2026

  Masking-based post-hoc explanation methods, such as KernelSHAP and LIME, estimate local feature importance by querying a black-box model under randomized perturbations. This paper formulates this procedure as communication over a query channel, where the latent explanation acts as a message and each…

• Strategic Decision Support for AI Agents

  Shayan Kiyani, Sima Noorani, George Pappas, Hamed Hassani · 12 juin 2026

  Traditionally, decision support studies how humans use machine learning models to make better decisions. In modern agentic systems, this division of roles is increasingly reversed: AI agents act on behalf of users, while humans and tools becomes support mechanisms around them. This role reversal bri…

• ATLAS: Active Theory Learning for Automated Science

  No\'emi \'Eltet\H{o}, Nathaniel D. Daw, Kimberly L. Stachenfeld, Kevin J. Miller · 11 juin 2026

  Advancing scientific understanding through mechanistic modeling requires posing the right experimental questions to yield maximally informative data. To automate this pursuit within cognitive science, we introduce ATLAS (Active Theory Learning for Automated Science), an active learning framework for…

• The Standard Interpretable Model: A general theory of interpretable machine learning to deductively design interpretable methods using Lagrangian mechanics

  Pietro Barbiero, Giovanni De Felice, Mateo Espinosa Zarlenga, Francesco Giannini, Filippo Bonchi, Mateja Jamnik, Giuseppe Marra, Ruggero Noris · 11 juin 2026

  As Artificial Intelligence models grow in complexity, interpretability has become an indispensable tool for understanding, debugging, and controlling their computations. However, interpretability lacks general theories to deductively design interpretable methods. This gap between theories and method…

• ICA Lens: Interpreting Language Models Without Training Another Dictionary

  Sida Liu, Feijiang Han · 11 juin 2026

  Finding interpretable directions in language-model representations is critical for understanding and controlling model behavior. Sparse autoencoders (SAEs) have become the standard tool for this purpose, but using them as the default first lens often requires training, storing, and evaluating large …

• When Context Returns: Toward Robust Internalization in On-Policy Distillation

  Xun Wang, Ruishuo Chen, Zhuoran Li, Yu Chen, Longbo Huang · 11 juin 2026

  Recent work has shown that on-policy distillation can internalize privileged context, such as system prompts or task hints, into a student model so that the context is no longer needed at inference time. Although this approach successfully improves the student's no-context performance, we identify a…

• Categorical Prior Lock-in: Why In-Context Learning Fails for Structured Data

  Antonio Pelusi, Stefano Braghin, Alberto Trombetta · 11 juin 2026

  Large language models (LLMs) are increasingly used as conditional generators for structured data, relying on in-context learning (ICL) to adapt to new distributions without parameter updates. We investigate the limits of ICL for structured generation under distribution mismatch, using high-cardinali…

• The Impossibility of Eliciting Latent Knowledge

  Korbinian Friedl, Francis Rhys Ward, Paul Yushin Rapoport, Tom Everitt, Jonathan Richens · 11 juin 2026

  Advanced AI systems have extensive knowledge of their environments; in fact, their knowledge may (far) exceed that of their developers or users. Consequently, a desirable property for an AI system is that it is honest -- that it accurately reports its beliefs about the world. Designing an AI system …

• When Do Data-Driven Systems Exhibit the Capability to Infer?

  Maximilian Poretschkin, Tabea Naeven · 11 juin 2026

  The European AI Act is the first comprehensive regulation of artificial intelligence (AI), setting out extensive obligations, particularly for so-called high-risk and general-purpose AI systems. A key distinguishing feature of AI systems under the AI Act is the capability to infer. Since the AI Act …

• StatefulDiscovery: Evidence-Calibrated Claim Formation in Open-Ended Scientific Discovery

  Jiayao Chen, Shi Liu, Linyi Yang · 11 juin 2026

  Open-ended scientific discovery asks agents to move beyond executing analyses for predefined questions. Across multiple rounds of exploration, a discovery agent must decide which phenomena warrant investigation while avoiding overinterpretation, where emerging claims exceed the evidential scope of t…

• From Consumption to Reflection: Designing Human-AI Relations for Stable Reasoning

  Rikard Rosenbacke, Carl Rosenbacke, Victor Rosenbacke, Martin McKee · 11 juin 2026

  Large language models (LLMs) have transformed how humans access information, but not how we reason with it. Their fluency accelerates consumption while bypassing the slow, reflective processes that underpin sound judgment. This paper introduces Relational Reflective Intelligence (RRI), an inference-…

• Calibration Drift Under Reasoning: How Chain-of-Thought Budgets Induce Overconfidence in Large Language Models

  Prakul Sunil Hiremath, Harshit R. Hiremath · 11 juin 2026

  The ability of large language models (LLMs) to express calibrated uncertainty is important for safe deployment. Chain-of-thought (CoT) reasoning is widely used to improve accuracy and reliability, but its effect on calibration is not fully understood. We show that this picture is incomplete: in some…