Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation

• URL: http://arxiv.org/abs/2606.19344v1
• Date: Fri, 24 Apr 2026 13:37:51 GMT
• Title: Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation
• Abstract summary: Large Language Models (LLMs) exhibit representational and syntactic biases that are difficult to evaluate due to the marginalization of text generation.<n>This paper introduces TreeTracer, a visual analytics tool designed to evaluate bias through aggregated comparison.
• Score: 16.63373250752545
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) exhibit representational and syntactic biases that are difficult to evaluate due to the stochastic nature of text generation. Standard auditing methods rely on a single output inspection or static automated metrics. These approaches obscure the underlying probability distributions and fail to capture biases hidden in lower-probability generation branches. This paper introduces TreeTracer, a visual analytics tool designed to evaluate LLM bias through aggregated comparison. Using a systematic perturbation analysis pipeline, the tool replaces ontology-defined terms in each input prompt, aggregates hundreds of stochastic generations into a syntax-aligned hierarchical structure, and then performs classification-aware node merging with an auxiliary language model. The resulting structure is visualized through a custom Sankey diagram. By juxtaposing two ontology-driven trees, the workspace enables direct comparison between semantic contexts and supports systematic bias detection. Because any visualization reflects only a subset of the model's learned behavior, the system further applies contrastive inference to compute and directly display counterfactual token probabilities across contexts, reducing the risk of misinterpreting the presence of bias. We validate the workspace through case studies comparing an unaligned baseline model GPT-2 XL against the constitutionally aligned Apertus models. The visual aggregation successfully exposes hidden representational harms, such as counterfactual pronoun suppression and conversational marginalization of individuals. A preliminary user study confirms that the aggregated comparative interface reduces cognitive load and effectively supports analysts in detecting systemic biases.

Related papers

• Dual-Branch Cross-Projection Debiasing through Diffusion-based Disentanglement [66.67463557274358]
  Foundation models trained on biased datasets often rely on spurious correlations between target labels and non-causal attributes.<n>We introduce Confidence-guided Concept Mining (CBCM) to identify reliable spurious attributes without attribute annotations.<n>We propose Dual-branch Cross-projection Debiasing (DCD), a prompt-tuning framework that separates target and spurious representations into two branches.
  arXiv  Detail & Related papers  (2026-06-23T05:28:47Z)
• Shorthand for Thought: Compressing LLM Reasoning via Entropy-Guided Supertokens [4.0704009036918025]
  Reasoning in Large Language Models incurs significant inference-time compute.<n>We show that reasoning tokens split into two functional types: low-entropy textitstructural tokens (recurring phrases that scaffold the reasoning process) and higher-entropy textitorganic tokens (problem-specific content that drives toward a solution)<n>We derive textitsupertokens that capture frequent structural patterns, then teach the model to adopt them via supervised fine-tuning.
  arXiv  Detail & Related papers  (2026-04-29T07:06:43Z)
• MuTSE: A Human-in-the-Loop Multi-use Text Simplification Evaluator [0.13048920509133807]
  textbfMuTSEfootnote is an interactive human-in-the-loop web application designed to streamline the evaluation of Large Language Models.<n>The system supports concurrent execution of $P times M$ prompt-model permutations, generating a comprehensive comparison matrix in real-time.
  arXiv  Detail & Related papers  (2026-04-10T04:31:43Z)
• UGID: Unified Graph Isomorphism for Debiasing Large Language Models [13.549042680933306]
  We propose underlineUnified underlineGraph underlineIsomorphism for underlineDebiasing large language models.<n>textittextbfUGID is an internal-representation-level debiasing framework for large language models.
  arXiv  Detail & Related papers  (2026-03-19T16:59:37Z)
• TIE: A Training-Inversion-Exclusion Framework for Visually Interpretable and Uncertainty-Guided Out-of-Distribution Detection [11.599035626374409]
  Deep neural networks often struggle to recognize when an input lies outside their training experience, leading to unreliable and overconfident predictions.<n>We propose textbfTIE: a Training--Inversion--Exclusion framework for visually interpretable and uncertainty-guided anomaly detection.
  arXiv  Detail & Related papers  (2025-11-28T22:06:01Z)
• SCALEX: Scalable Concept and Latent Exploration for Diffusion Models [59.86284983662119]
  Image generation models frequently encode social biases, including stereotypes tied to gender, race, and profession.<n>We introduce SCALEX, a framework for scalable and automated exploration of diffusion model latent spaces.<n>It extracts semantically meaningful directions from H-space using only natural language prompts, enabling zero-shot interpretation without retraining or labelling.
  arXiv  Detail & Related papers  (2025-11-13T22:02:44Z)
• Mitigating Multimodal Hallucinations via Gradient-based Self-Reflection [49.26064449816502]
  We propose a Gradient-based Influence-Aware Constrained Decoding (GACD) method to address text-visual bias and co-occurrence bias.<n>GACD effectively reduces hallucinations and improves the visual grounding of MLLM outputs.
  arXiv  Detail & Related papers  (2025-09-03T08:13:52Z)
• Semantic and Structural Analysis of Implicit Biases in Large Language Models: An Interpretable Approach [1.5749416770494704]
  It proposes an interpretable bias detection method aimed at identifying hidden social biases in model outputs.<n>The method combines nested semantic representation with a contextual contrast mechanism.<n>The evaluation focuses on several key metrics, such as bias detection accuracy, semantic consistency, and contextual sensitivity.
  arXiv  Detail & Related papers  (2025-08-08T09:21:10Z)
• Measuring the Interpretability of Unsupervised Representations via Quantized Reverse Probing [97.70862116338554]
  We investigate the problem of measuring interpretability of self-supervised representations. We formulate the latter as estimating the mutual information between the representation and a space of manually labelled concepts. We use our method to evaluate a large number of self-supervised representations, ranking them by interpretability.
  arXiv  Detail & Related papers  (2022-09-07T16:18:50Z)
• Causal Transportability for Visual Recognition [70.13627281087325]
  We show that standard classifiers fail because the association between images and labels is not transportable across settings. We then show that the causal effect, which severs all sources of confounding, remains invariant across domains. This motivates us to develop an algorithm to estimate the causal effect for image classification.
  arXiv  Detail & Related papers  (2022-04-26T15:02:11Z)
• Visualizing Classifier Adjacency Relations: A Case Study in Speaker Verification and Voice Anti-Spoofing [72.4445825335561]
  We propose a simple method to derive 2D representation from detection scores produced by an arbitrary set of binary classifiers. Based upon rank correlations, our method facilitates a visual comparison of classifiers with arbitrary scores. While the approach is fully versatile and can be applied to any detection task, we demonstrate the method using scores produced by automatic speaker verification and voice anti-spoofing systems.
  arXiv  Detail & Related papers  (2021-06-11T13:03:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.