A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Large Language Models

• URL: http://arxiv.org/abs/2601.17952v1
• Date: Sun, 25 Jan 2026 19:03:04 GMT
• Title: A Monosemantic Attribution Framework for Stable Interpretability in Clinical Neuroscience Large Language Models
• Authors: Michail Mamalakis, Tiago Azevedo, Cristian Cosentino, Chiara D'Ercoli, Subati Abulikemu, Zhongtian Sun, Richard Bethlehem, Pietro Lio,
• Abstract summary: Interpretability remains a key challenge for deploying large language models (LLMs) in clinical settings such as Alzheimer's disease progression diagnosis.<n>We introduce a unified interpretability framework that integrates attributional and mechanistic perspectives.
• Score: 9.694820939059339
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Interpretability remains a key challenge for deploying large language models (LLMs) in clinical settings such as Alzheimer's disease progression diagnosis, where early and trustworthy predictions are essential. Existing attribution methods exhibit high inter-method variability and unstable explanations due to the polysemantic nature of LLM representations, while mechanistic interpretability approaches lack direct alignment with model inputs and outputs and do not provide explicit importance scores. We introduce a unified interpretability framework that integrates attributional and mechanistic perspectives through monosemantic feature extraction. By constructing a monosemantic embedding space at the level of an LLM layer and optimizing the framework to explicitly reduce inter-method variability, our approach produces stable input-level importance scores and highlights salient features via a decompressed representation of the layer of interest, advancing the safe and trustworthy application of LLMs in cognitive health and neurodegenerative disease.

Related papers

• Explainable Parkinsons Disease Gait Recognition Using Multimodal RGB-D Fusion and Large Language Models [6.2676602262188625]
  This paper presents an explainable multimodal framework that integrates RGB and Depth (RGB-D) data to recognize Parkinsonian gait patterns.<n>By combining multimodal feature learning with language-based interpretability, this study bridges the gap between visual recognition and clinical understanding.
  arXiv  Detail & Related papers  (2025-12-04T03:43:43Z)
• nnMIL: A generalizable multiple instance learning framework for computational pathology [11.640858438464159]
  nnMIL is a learning framework that connects patch-level foundation models to robust slide-level clinical inference.<n>nnMIL consistently outperformed existing MIL methods for disease diagnosis, histologic subtyping, molecular biomarker detection, and pan- cancer prognosis prediction.<n>In conclusion, nnMIL offers a practical and generalizable solution for translating pathology foundation models into clinically meaningful predictions.
  arXiv  Detail & Related papers  (2025-11-18T20:56:37Z)
• LTD-Bench: Evaluating Large Language Models by Letting Them Draw [57.237152905238084]
  LTD-Bench is a breakthrough benchmark for large language models (LLMs)<n>It transforms LLM evaluation from abstract scores to directly observable visual outputs by requiring models to generate drawings through dot matrices or executable code.<n> LTD-Bench's visual outputs enable powerful diagnostic analysis, offering a potential approach to investigate model similarity.
  arXiv  Detail & Related papers  (2025-11-04T08:11:23Z)
• CANDLE: A Cross-Modal Agentic Knowledge Distillation Framework for Interpretable Sarcopenia Diagnosis [3.0245458192729466]
  CANDLE mitigates the interpretability-performance trade-off, enhances predictive accuracy, and preserves high decision consistency.<n>The framework offers a scalable approach to knowledge assetization of TML models, enabling interpretable, reproducible, and clinically aligned decision support in sarcopenia and potentially broader medical domains.
  arXiv  Detail & Related papers  (2025-07-26T15:50:08Z)
• MIRROR: Multi-Modal Pathological Self-Supervised Representation Learning via Modality Alignment and Retention [57.044719143401664]
  Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease.<n>We present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention.<n>Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance.
  arXiv  Detail & Related papers  (2025-03-01T07:02:30Z)
• LlaMADRS: Prompting Large Language Models for Interview-Based Depression Assessment [75.44934940580112]
  This study introduces LlaMADRS, a novel framework leveraging open-source Large Language Models (LLMs) to automate depression severity assessment.<n>We employ a zero-shot prompting strategy with carefully designed cues to guide the model in interpreting and scoring transcribed clinical interviews.<n>Our approach, tested on 236 real-world interviews, demonstrates strong correlations with clinician assessments.
  arXiv  Detail & Related papers  (2025-01-07T08:49:04Z)
• Knowledge-Augmented Multimodal Clinical Rationale Generation for Disease Diagnosis with Small Language Models [14.136585695164426]
  Small language models (SLMs) are efficient but lack advanced reasoning for integrating multimodal medical data.<n>We propose ClinRaGen, enhancing SLMs by leveraging LLM-derived reasoning ability via rationale distillation and domain knowledge injection.<n> Experiments on real-world medical datasets show that ClinRaGen achieves state-of-the-art performance in disease diagnosis and rationale generation.
  arXiv  Detail & Related papers  (2024-11-12T07:34:56Z)
• Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness [68.69369585600698]
  Deep learning models often suffer from a lack of interpretability due to polysemanticity. Recent advances in monosemanticity, where neurons correspond to consistent and distinct semantics, have significantly improved interpretability. We show that monosemantic features not only enhance interpretability but also bring concrete gains in model performance.
  arXiv  Detail & Related papers  (2024-10-27T18:03:20Z)
• Kernel Language Entropy: Fine-grained Uncertainty Quantification for LLMs from Semantic Similarities [79.9629927171974]
  Uncertainty in Large Language Models (LLMs) is crucial for applications where safety and reliability are important. We propose Kernel Language Entropy (KLE), a novel method for uncertainty estimation in white- and black-box LLMs.
  arXiv  Detail & Related papers  (2024-05-30T12:42:05Z)
• Beyond Self-Consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Staging [0.33554367023486936]
  Cancer staging status is available in clinical reports, but it requires natural language processing to extract it. With the advance in clinical-oriented large language models, it is promising to extract such status without extensive efforts in training the algorithms. In this study, we propose an ensemble reasoning approach with the aim of improving the consistency of the model generations.
  arXiv  Detail & Related papers  (2024-04-19T19:34:35Z)
• Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention [53.896974148579346]
  Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains. The enigmatic black-box'' nature of LLMs remains a significant challenge for interpretability, hampering transparent and accountable applications. We propose a novel methodology anchored in sparsity-guided techniques, aiming to provide a holistic interpretation of LLMs.
  arXiv  Detail & Related papers  (2023-12-22T19:55:58Z)

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.