Beyond Endpoint Gains: A Weight-Delta Audit of Medical Specialization

• URL: http://arxiv.org/abs/2608.20768v3
• Date: Thu, 27 Aug 2026 05:28:17 GMT
• Title: Beyond Endpoint Gains: A Weight-Delta Audit of Medical Specialization
• Abstract summary: We propose a paired weight-delta path audit and apply it to two public, aligned generalist-to-medical-specialist checkpoint pairs.<n>Its claims concern text-only multiple-choice benchmark movement, not clinical validation, repair, or circuit-level mechanism.
• Score: 2.782879873824998
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Specialist language models are usually understood through endpoint gains: the generalist scores lower, the specialist scores higher, and the difference is treated as evidence of specialization. This leaves the released update itself largely unexamined. We propose a paired weight-delta path audit and apply it to two public, aligned generalist-to-medical-specialist checkpoint pairs: Gemma-3-4B-IT to MedGemma-4B-IT and Qwen2.5-7B-Instruct to HuatuoGPT-o1-7B. In both pairs, the full decoder-side update strongly reconstructs measured medical benchmark movement (0.974 and 1.183 endpoint-normalized retention), making each decoder delta an appropriate substrate for the audit. Yet the movement is not cleanly localized. MLP is the strongest broad component family in both pairs, but mixed off-domain movements, 10-seed matched controls, and endpoint-anchored rollbacks prevent a unique coarse-family explanation. The audit therefore separates update-level reconstruction from component-level explanation. Its claims concern text-only multiple-choice benchmark movement, not clinical validation, repair, or circuit-level mechanism.

Related papers

• MedSNIP: Building and Benchmarking Snippet-Level Granularity for Medical Fact Verification [50.13873960887466]
  We introduce MedSNIP-Bench, a benchmark for snippet-level medical fact verification.<n>We also introduce MedSNIP, an automatic snippet-generation pipeline.<n>Across MedSNIP-Bench, HealthFC, and MedHallu, snippet-level verification preserves or improves false-class F1.
  arXiv  Detail & Related papers  (2026-09-11T14:08:44Z)
• Differentiable Interval Bottlenecks for Interpretable Anomaly Detection in Numerical Data [0.5213778368155993]
  DIFFINT is an autoencoder whose latent bottleneck is structured as a set of soft, axis-aligned interval memberships.<n>It is the only interpretable detector in the statistically-tied leading cluster of seven methods.
  arXiv  Detail & Related papers  (2026-09-03T14:05:27Z)
• Loud or Silent? A Reusable Framework for Per-Modality Failure Analysis in Multimodal Clinical AI [3.7524666944787644]
  Two questions matter beyond the size of the accuracy loss: which modality was responsible, and whether the model fails loudly or silently once that modality is dropped.<n>We present a model-agnostic modality-failure framework: given N modality embeddings, any mask-aware probe, and labels, it returns a per-example failure taxonomy.<n>It reports per-modality loud-vs-silent rates, and scales to a three-modality complementarity matrix.
  arXiv  Detail & Related papers  (2026-08-02T19:38:56Z)
• Multimodal Domain Generalization for Depression Detection: An Attention-Based BiLSTM Network with Domain-Adversarial Training [4.970364068620607]
  We present the first patient-independent multimodal depression detection framework that incorporates domain generalization (DG)<n>The proposed model integrates bidirectional Long Short-Term Memory (BiLSTM) with intra- and cross-modal attention mechanisms, accompanied by segment-level fusion for decision-making.<n>The addition of DG to this baseline yields a 2.5% increase in accuracy and 3.3% in F1-score, achieving 93.2% accuracy, 93.2% precision, 96.2% recall, and 94.2% F1-score, surpassing all existing benchmarks.
  arXiv  Detail & Related papers  (2026-07-24T13:52:27Z)
• Fantastic Adaptive Taxonomies and How to Use Them [100.20614173157708]
  AdaMAST converts execution traces into a compact, evidence-grounded failure taxonomy.<n>No code is hand-authored, no trace human-annotated.<n>In agent-system search, taxonomy-coded diagnoses of failed candidates outperform free-form.
  arXiv  Detail & Related papers  (2026-07-17T17:41:16Z)
• Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems [47.027290803102666]
  Normal behaviour is the union of many imbalanced, curved, thin-fringed operating regimes.<n>We model the normal law with a jointly learned latent representation plus explicit Gaussian-mixture mode clustering.<n>We evaluate under a deliberately fair protocol: raw point-wise metrics with no point adjustment, a trivial-detector difficulty split, prevalence-matched F1, and train-normal-only calibration.
  arXiv  Detail & Related papers  (2026-07-07T10:10:00Z)
• Agentic AI-based Framework for Mitigating Premature Diagnostic Handoff and Silent Hallucination in Healthcare Applications [17.69274632062373]
  Open-ended conversational agents remain prone to two critical failure modes: premature diagnostic handoff and silent clinical hallucinations.<n>We propose a multi-agent framework that addresses both issues by replacing LLM-as-a-judge'' routing with deterministic orchestration constraints.<n>We evaluate the system using simulated patient agents powered by the llama-3.1-70b-instruct model on 150 test cases.
  arXiv  Detail & Related papers  (2026-06-16T15:39:19Z)
• Shift-Dependent Asymmetry: Orthogonal Inverse Low-Rank Adaptation for Federated Medical Segmentation [60.41721951405564]
  Low-Rank Adaptation (LoRA) enables efficient federated fine-tuning of segmentation foundation models for medical imaging.<n>Most federated LoRA methods adopt a uniform aggregation rule, which breaks under the encoder-decoder asymmetry in medical segmentation.<n>This mismatch entangles shared anatomy with site-specific biases and harms generalization.<n>We introduce a Subspace Orthogonality Regularizer that penalizes shared-local collinearity in the effective update space.
  arXiv  Detail & Related papers  (2026-06-07T15:39:28Z)
• GenGait: A Transformer-Based Model for Human Gait Anomaly Detection and Normative Twin Generation [11.553495883461359]
  This work proposes a label-free framework for joint-level anomaly kinematic correction based on a Transformer masked autoencoder.<n>The proposed approach enables interpretable, subject-specific localization of gait impairments without requiring disease labels.
  arXiv  Detail & Related papers  (2026-04-02T13:06:27Z)
• Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning [53.45095336430027]
  We develop a unified framework that combines implicit retrieval and structured collaboration.<n>On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3% accuracy.<n>Results on SuperGPQA and TRQA confirm robustness across domains.
  arXiv  Detail & Related papers  (2025-09-25T14:05:55Z)
• Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models [57.20761595019967]
  We present Normalized Attention Guidance (NAG), an efficient, training-free mechanism that applies extrapolation in attention space with L1-based normalization and refinement.<n>NAG restores effective negative guidance where CFG collapses while maintaining fidelity.<n>NAG generalizes across architectures (UNet, DiT), sampling regimes (few-step, multi-step), and modalities (image, video)
  arXiv  Detail & Related papers  (2025-05-27T13:30:46Z)
• TAGS: A Test-Time Generalist-Specialist Framework with Retrieval-Augmented Reasoning and Verification [27.918192794935255]
  We present TAGS, a test-time framework that combines a broadly capable exemplar with a domain-specific specialist to offer complementary perspectives.<n>TAGS achieves strong performance across nine MedQA benchmarks, boosting GPT-4o accuracy by 13.8%, DeepSeek-R1 by 16.8%, and improving a vanilla 7B model from 14.1% to 23.9%.
  arXiv  Detail & Related papers  (2025-05-23T18:28:59Z)

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.