arXiv subjects

Chris Russell

Publications and source records attributed to Chris Russell.

At least 19 recordsLinked to original sources

arXiv2026

Rule of Thumb: Explaining Artificial Intelligence Systems using Partial Information

Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ''Rule of Thumb'' (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI in: (a) zero-shot classification using large language models (LLMs), (b) auditing of opaque AI systems without model access, and (c) the use of AI in scientific discovery. Additionally, RoT meets specific requirements from leading AI regulations, provides a familiar interface and visualisations for XAI practitioners, is model-agnostic, and is substantially faster than alternatives. Code available at: https://github.com/KaiRawal/Rule-of-Thumb-Explaining-Artificial-Intelligence-Systems-using-Partial-Information

cs.AI↗

arXiv2026

A Technical Typology of AI Systems in Public Administration

Research on artificial intelligence (AI) in the public sector often treats "AI" as a single category, neglecting technical distinctions between different AI systems. But these distinctions affect how different systems impact core public values like accountability, procedural justice, and non-discrimination. This paper argues that public administration research would benefit from more technical precision on "AI" and makes three contributions to this end. First, we introduce a typology of five categories of AI systems: hand-coded, glass-box, black-box, general-purpose, and agentic systems. We calibrate the typology to public administration by grouping system types by their distinct implications for public values. Second, we evaluate technical precision in recent public administration research about AI by coding 91 highly-cited papers (2019-2025) using our typology. We find widespread imprecision: most papers (55\%) leave the studied system underspecified, 31\% motivate their work with a different system than they study, and 41\% make more general conclusions than the studied system supports. Finally, we give practical recommendations for future research. We highlight common pitfalls to avoid, and suggest that researchers should, at a minimum, provide enough technical detail to locate the studied system in our typology. To this end, we provide a practical guide -- a short set of diagnostic questions answerable from public information and without specialist technical knowledge.

cs.CY↗

arXiv2026

The Heterogeneous Safety Impacts of Benign Multilingual Fine-Tuning

Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task. However, prior work has shown that this increase in capability comes with a cost: it can increase a model's tendency to respond to unsafe adversarial prompts, even when fine-tuning with non-adversarial data. We present the first comprehensive empirical study of this phenomenon in multilingual settings by fine-tuning Llama-3.2, Qwen3, and Gemma-3 models using benign data translated across nine languages. We find that safety outcomes are highly sensitive to both the choice of fine-tuning language and the evaluation language, with adversarial compliance rates increasing four-fold in some settings. Multilingual safety drift is decoupled from general capability metrics, and occurs heterogeneously across languages and models. Fine-tuning in non-English languages often induces smaller internal representational drifts than English, but these shifts lead models to default to either exaggerated compliance or refusal. As such, assessing fine-tuning impacts solely in English provides inadequate assurance for deployment. To facilitate further research into these cross-lingual safety blind spots, we release the Multilingual-Benign-Tune dataset and the SORRY-Bench-Multilingual evaluation suite.

cs.CL↗

arXiv2026

Running the Gauntlet: Re-evaluating the Capabilities of Agents Beyond Familiar Environments

As agentic systems continue to evolve and are widely deployed in real-world scenarios, there is a growing demand to faithfully evaluate their capabilities. However, current benchmarks are typically built on popular applications with relatively simple tasks and focus on a narrow set of capabilities while overlooking broader dimensions, resulting in saturated performance on modern agents and failing to probe their limitations. To this end, we introduce GauntletBench, a web-based benchmark for evaluating agent generalisation in challenging scenarios, focusing on three underexplored capabilities (temporal perception, graphical understanding, and 3D reasoning), across five less-covered professional applications (Video Editor, Workflow Builder, 3D Modeller, Flight Analyser, and Circuit Designer), each with 27 vision-intensive tasks (135 in total). Our benchmark provides a modular pipeline that comprises an environment compatible with both open- and closed-source agent frameworks, a controlled web-based application, a well-structured task suite, and an automated evaluation engine with diverse metrics. Contrary to widespread expectations, our empirical results reveal that frontier agentic systems remain far from achieving human-level performance. Even the state-of-the-art agent achieves only a 28.2% success rate on our GauntletBench, highlighting the limitations in these overlooked capabilities and generalisation. By comparison, non-expert human annotators achieve over 80% success on our challenging yet feasible tasks, revealing the substantial gap between current agent capabilities and those required for complex real-world scenarios.

cs.LG↗

arXiv2026

Adaptive Calibration for Fair and Performant Facial Recognition

We introduce Adaptive Calibration (AC), a novel calibration strategy for facial recognition that maps cosine similarity between normalized embeddings to well-calibrated probabilities. By incorporating local context into calibration, Adaptive Calibration corrects for a fundamental mismatch in cosine similarity, whereby the same distance can correspond to different match probabilities in different embedding regions. Our approach improves both overall performance and results in a fairer calibration without requiring demographic metadata. Our approach often improves worst-group ranking and probability quality across a variety of pretrained models and standard benchmarks, with gains in low-FPR verification performance depending on the setting. AC provides a practical solution for equitable facial recognition, without requiring demographic group annotations, and while improving overall performance. Our method provides continuous, region-specific calibration, and we examine per-group performance to assess "leveling down" where fairness comes at the cost of degraded performance for some groups.

cs.CV↗

arXiv2026

Toward Calibrated, Fair, and accurate Deepfake Detection

Deepfake detectors show large performance gaps across demographic groups. Existing fairness approaches require demographic labels, retraining, or sacrifice accuracy. We introduce Face-Fairness (FF), a plug-and-play framework for bias mitigation. Our primary contribution, Face-Feature Tuning (FFT), is the first demographic label-free fairness method demonstrated for deepfake detection: a lightweight calibrator that performs a logit remapping conditioned on frozen face embeddings. We complement FFT with two variants: FF-Max, which maximizes worst-group accuracy when demographics are available, and FF-Discover, which does the same with embedding-discovered groups. Across in-domain and cross-dataset test settings, FF consistently reduces FPR/TPR gaps and improves minimum group accuracy while maintaining (often improving) overall accuracy. The approach is detector-agnostic, adds negligible runtime overhead, and requires no access to identity attributes.

cs.LG↗

arXiv2026

Grounding Text Embeddings in Stakeholder Associations

Text embeddings are widely used to analyse large corpora of complex texts. However, it is unclear whether the embeddings capture the same semantic distances as the human experts using them. Ensuring alignment between embedding representations and human intentions is essential for valid analyses. We present the Stakeholder Grounding Exercise, a method for making expert associations explicit and grounding embedding model results in human understanding. In our primary case study on Danish policy issues, we find that neural text embeddings are substantially less reliable than human experts (19-26 pp gap), and that this misalignment propagates to downstream clustering performance (Spearman $\rho=0.9$ between exercise ranking and cluster quality). A secondary study on US Federal AI use cases replicates the gap (16pp) in English, using a digital protocol and a different community of experts -- demonstrating that the gap is not an artefact of a single instrument or domain. The Stakeholder Grounding Exercise offers a practical method for assessing whether embedding models capture the semantic distinctions that matter most to domain experts.

cs.CL↗

arXiv2026

AI-Mediated Communication Can Steer Collective Opinion

Generative artificial intelligence (AI) is increasingly integrated into the online platforms where humans exchange opinions; large language models (LLMs) now polish users' posts on LinkedIn and provide context for content shared on X. While prior work has shown that AI can express biased opinions and shape individuals' opinions during human-AI interactions, less attention has been paid to its influence on collective opinion formation when mediating human-to-human communication. We address this gap via a combination of empirical and theoretical analyses. We show empirically that LLMs from multiple popular families introduce directional biases when instructed to edit human-written texts on contested topics, for example, nudging texts in favor of gun control and against atheism. Building on this observation, we introduce a mathematical model of opinion dynamics in which an AI system sits between users on a social network, transforming the opinions they express and perceive. By analytically characterizing the equilibrium of this model and performing simulations on real social network data, we show that biases introduced by AI in human-to-human communication can be amplified through the network and shift collective opinion in their direction. In light of these findings, we investigate whether such biases are controllable by online platforms. We audit the "Explain this post" feature on X and find evidence of pro-life bias in Grok's outputs on abortion-related content, which we trace back to specific design choices. We conclude with a discussion of the broader implications of our findings in relation to ongoing legislative efforts in the European Union.

cs.CY↗

arXiv2026

Known By Their Actions: Fingerprinting LLM Browser Agents via UI Traces

As LLM-based agents increasingly browse the web on users' behalf, a natural question arises: can websites passively identify which underlying model powers an agent? Doing so would represent a significant security risk, enabling targeted attacks tailored to known model vulnerabilities. Across 14 frontier LLMs and four web environments spanning information retrieval and shopping tasks, we show that an agent's actions and interaction timings, captured via a passive JavaScript tracker, are sufficient to identify the underlying model with up to 96\% F1. We formalise this attack surface by demonstrating that classifiers trained on agent actions generalise across model sizes and families. We further show that strong classifiers can be trained from few interaction traces and that agent identity can be inferred early within an episode. Injecting randomised timing delays between actions substantially degrades classifier performance, but does not provide robust protection: a classifier retrained on delayed traces largely recovers performance. We release our harness and a labelled corpus of agent traces \href{https://github.com/KabakaWilliam/known_actions}{here}.

cs.CR↗

arXiv2026

Task-Specific Knowledge Distillation via Intermediate Probes

Knowledge distillation from large language models (LLMs) assumes that the teacher's output distribution is a high-quality training signal. On reasoning tasks, this assumption is frequently violated. A model's intermediate representations may encode the correct answer, yet this information is lost or distorted through the vocabulary projection, where prompt formatting and answer-token choices creates brittle, noisy outputs. We introduce \method{}, a distillation framework that bypasses this bottleneck by training lightweight probes on frozen teacher hidden states and using the probe's predictions, rather than output logits, as supervision for student training. This simple change yields consistent improvements across four reasoning benchmarks (AQuA-RAT, ARC Easy/Challenge, and MMLU), with gains most pronounced under limited data. Probes trained on intermediate representations provide cleaner labels than the teacher's own outputs, effectively denoising the distillation signal. \method{} requires no architectural changes to student or teacher, is architecture-agnostic, and adds minimal compute since probe training is cheap and teacher representations can be cached. By exploiting internal representations, \method{} enables practitioners to extract more value from large teacher models without additional training data or architectural complexity.

cs.CL↗

arXiv2026

LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations

Running LLMs with extended reasoning on every problem is expensive, but determining which inputs actually require additional compute remains challenging. We investigate whether their own likelihood of success is recoverable from their internal representations before generation, and if this signal can guide more efficient inference. We train linear probes on pre-generation activations to predict policy-specific success on math and coding tasks, substantially outperforming surface features such as question length and TF-IDF. Using E2H-AMC, which provides both human and model performance on identical problems, we show that models encode a model-specific notion of difficulty that is distinct from human difficulty, and that this distinction increases with extended reasoning. Leveraging these probes, we demonstrate that routing queries across a pool of models can exceed the best-performing model whilst reducing inference cost by up to 70\% on MATH, showing that internal representations enable practical efficiency gains even when they diverge from human intuitions about difficulty. Our code is available at: https://github.com/KabakaWilliam/llms_know_difficulty

cs.CL↗

arXiv2026

Agent Benchmarks Fail Public Sector Requirements

Deploying Large Language Model-based agents (LLM agents) in the public sector requires assuring that they meet the stringent legal, procedural, and structural requirements of public-sector institutions. Practitioners and researchers often turn to benchmarks for such assessments. However, it remains unclear what criteria benchmarks must meet to ensure they adequately reflect public-sector requirements, or how many existing benchmarks do so. In this paper, we first define such criteria based on a first-principles survey of public administration literature: benchmarks must be \emph{process-based}, \emph{realistic}, \emph{public-sector-specific} and report \emph{metrics} that reflect the unique requirements of the public sector. We analyse more than 1,300 benchmark papers for these criteria using an expert-validated LLM-assisted pipeline. Our results show that no single benchmark meets all of the criteria. Our findings provide a call to action for both researchers to develop public sector-relevant benchmarks and for public-sector officials to apply these criteria when evaluating their own agentic use cases.

cs.CY↗

arXiv2026

Evaluating the Ability of Explanations to Disambiguate Models in a Rashomon Set

Explainable artificial intelligence (XAI) is concerned with producing explanations indicating the inner workings of models. For a Rashomon set of similarly performing models, explanations provide a way of disambiguating the behavior of individual models, helping select models for deployment. However explanations themselves can vary depending on the explainer used, and need to be evaluated. In the paper "Evaluating Model Explanations without Ground Truth", we proposed three principles of explanation evaluation and a new method "AXE" to evaluate the quality of feature-importance explanations. We go on to illustrate how evaluation metrics that rely on comparing model explanations against ideal ground truth explanations obscure behavioral differences within a Rashomon set. Explanation evaluation aligned with our proposed principles would highlight these differences instead, helping select models from the Rashomon set. The selection of alternate models from the Rashomon set can maintain identical predictions but mislead explainers into generating false explanations, and mislead evaluation methods into considering the false explanations to be of high quality. AXE, our proposed explanation evaluation method, can detect this adversarial fairwashing of explanations with a 100% success rate. Unlike prior explanation evaluation strategies such as those based on model sensitivity or ground truth comparison, AXE can determine when protected attributes are used to make predictions.

cs.AI↗

arXiv2025

It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents

Web-based agents powered by large language models are increasingly used for tasks such as email management or professional networking. Their reliance on dynamic web content, however, makes them vulnerable to prompt injection attacks: adversarial instructions hidden in interface elements that persuade the agent to divert from its original task. We introduce the Task-Redirecting Agent Persuasion Benchmark (TRAP), a benchmark for studying how persuasion techniques misguide autonomous web agents on realistic tasks. Across six frontier models, agents are susceptible to prompt injection in 25% of tasks on average (13% for GPT-5 to 43% for DeepSeek-R1), with small interface or contextual changes often doubling success rates and revealing systemic, psychologically driven vulnerabilities in web-based agents. We also provide a modular social-engineering injection framework with controlled experiments on high-fidelity website clones, allowing for further benchmark expansion.

cs.HC↗

arXiv2025

OxEnsemble: Fair Ensembles for Low-Data Classification

We address the problem of fair classification in settings where data is scarce and unbalanced across demographic groups. Such low-data regimes are common in domains like medical imaging, where false negatives can have fatal consequences. We propose a novel approach \emph{OxEnsemble} for efficiently training ensembles and enforcing fairness in these low-data regimes. Unlike other approaches, we aggregate predictions across ensemble members, each trained to satisfy fairness constraints. By construction, \emph{OxEnsemble} is both data-efficient -- carefully reusing held-out data to enforce fairness reliably -- and compute-efficient, requiring little more compute than used to fine-tune or evaluate an existing model. We validate this approach with new theoretical guarantees. Experimentally, our approach yields more consistent outcomes and stronger fairness-accuracy trade-offs than existing methods across multiple challenging medical imaging classification datasets.

cs.CV↗

arXiv2025

FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models

Text-to-image diffusion models, such as Stable Diffusion, have demonstrated remarkable capabilities in generating high-quality and diverse images from natural language prompts. However, recent studies reveal that these models often replicate and amplify societal biases, particularly along demographic attributes like gender and race. In this paper, we introduce FairImagen (https://github.com/fuzihaofzh/FairImagen), a post-hoc debiasing framework that operates on prompt embeddings to mitigate such biases without retraining or modifying the underlying diffusion model. Our method integrates Fair Principal Component Analysis to project CLIP-based input embeddings into a subspace that minimizes group-specific information while preserving semantic content. We further enhance debiasing effectiveness through empirical noise injection and propose a unified cross-demographic projection method that enables simultaneous debiasing across multiple demographic attributes. Extensive experiments across gender, race, and intersectional settings demonstrate that FairImagen significantly improves fairness with a moderate trade-off in image quality and prompt fidelity. Our framework outperforms existing post-hoc methods and offers a simple, scalable, and model-agnostic solution for equitable text-to-image generation.

cs.LG↗

arXiv2025

LLMs Encode How Difficult Problems Are

Large language models exhibit a puzzling inconsistency: they solve complex problems yet frequently fail on seemingly simpler ones. We investigate whether LLMs internally encode problem difficulty in a way that aligns with human judgment, and whether this representation tracks generalization during reinforcement learning post-training. We train linear probes across layers and token positions on 60 models, evaluating on mathematical and coding subsets of Easy2HardBench. We find that human-labeled difficulty is strongly linearly decodable (AMC: $\rho \approx 0.88$) and exhibits clear model-size scaling, whereas LLM-derived difficulty is substantially weaker and scales poorly. Steering along the difficulty direction reveals that pushing models toward "easier" representations reduces hallucination and improves accuracy. During GRPO training on Qwen2.5-Math-1.5B, the human-difficulty probe strengthens and positively correlates with test accuracy across training steps, while the LLM-difficulty probe degrades and negatively correlates with performance. These results suggest that human annotations provide a stable difficulty signal that RL amplifies, while automated difficulty estimates derived from model performance become misaligned precisely as models improve. We release probe code and evaluation scripts to facilitate replication.

cs.CL↗

arXiv2025

CAST: Compositional Analysis via Spectral Tracking for Understanding Transformer Layer Functions

Large language models have achieved remarkable success but remain largely black boxes with poorly understood internal mechanisms. To address this limitation, many researchers have proposed various interpretability methods including mechanistic analysis, probing classifiers, and activation visualization, each providing valuable insights from different perspectives. Building upon this rich landscape of complementary approaches, we introduce CAST (Compositional Analysis via Spectral Tracking), a probe-free framework that contributes a novel perspective by analyzing transformer layer functions through direct transformation matrix estimation and comprehensive spectral analysis. CAST offers complementary insights to existing methods by estimating the realized transformation matrices for each layer using Moore-Penrose pseudoinverse and applying spectral analysis with six interpretable metrics characterizing layer behavior. Our analysis reveals distinct behaviors between encoder-only and decoder-only models, with decoder models exhibiting compression-expansion cycles while encoder models maintain consistent high-rank processing. Kernel analysis further demonstrates functional relationship patterns between layers, with CKA similarity matrices clearly partitioning layers into three phases: feature extraction, compression, and specialization.

cs.LG↗