Medical
-
Semantic Latent Alignment for Flow Matching with Invertible Bi-Lipschitz Adapters
2026-10-15 06:19:41Felix N\"utzel, Mischa Dombrowski, Luca Hagen, Bernhard KainzAbstract
Pretrained latent spaces trade compression, reconstruction, and the learnability of the coordinates a downstream generator has to work in, which complicates adapting these spaces to different domains or tasks: continued training is expensive, and without a new alignment objective it often fails to improve downstream generation. We instead leave the encoder and decoder fixed and change only the coordinates between them, using an invertible bi-Lipschitz adapter aligned offline to a semantic teacher and frozen during class-conditional flow matching. Invertibility preserves compatibility with the pretrained autoencoder, while the bi-Lipschitz constraint limits how far alignment can distort its latent geometry. This provides a simple way to improve semantic organization while retaining the compression and reconstruction properties of the original latent space. On two medical imaging datasets, constrained latent adaptation improves feature metrics, seed stability and counterfactual edit strength, while remaining competitive on downstream classification. In contrast, unconstrained adaptation can become highly anisotropic and sensitive despite remaining invertible, highlighting the importance of controlling geometric distortion. Our results suggest that carefully constrained post-hoc adaptation offers a parameter-efficient and tuning-free alternative to jointly modifying the representation during generative training.
Abstract (translated)
URL
https://arxiv.org/abs/2610.21332
PDF
-
ZoomCap: Evidence-Routed Visual Zooming for Recoverable Visual Information Transfer
2026-10-15 06:10:24Hao Shi, Jiafeng Wang, Yuzhuo Fu, Tao Zhang, Jian Bai, Juelu Zhang, Qing Cui, Chao HuangAbstract
Detailed captions transfer visual information into vision-language model supervision, but quality is easily conflated with length: long single-pass outputs still miss small objects, text, and spatial relations or promote uncertain observations into confident claims. We formulate fine-grained captioning as recoverable visual information transfer, in which a teacher acquires missing evidence under a finite observation budget and converts it into reusable supervision. ZoomCap uses a draft caption to route conditional knowledge retrieval, OCR-sensitive reading, coverage-gap recovery, and adaptive visual zooming, allocating deeper crops only to the most complex eligible region, deriving spatial relations from grounded geometry, and fusing non-duplicate evidence. We also introduce FactCycle, which scores caption-derived atomic claims on two blind anchors--source image (truthfulness) and reconstruction (realizability)--separating fabrication from generator loss claim by claim; human and multi-judge comparisons complement this fact-level view. Against Qwen3.5-397B-A17B captions, ZoomCap achieves an 81.9% human tie-adjusted preference score with a teacher over 10x smaller, and FactCycle counts 52.4 transferred claims per image vs. 31.5 for ScaleCap. The evidence also transfers: in continual pre-training, ZoomCap-450K improves the 11-benchmark average by 1.31-2.38 points over the strongest competing corpus and Ling-Tiny-VL-v3 gains 1.4 points on a production document/OCR monitor suite (1.5 on perception) at 60B tokens; in medical pre-training, 2.64M ZoomCap-captioned images improve five VQA benchmarks by 3.04 points (vs. 1.76 for same-image Gemini-3.1-Pro captions).
Abstract (translated)
URL
https://arxiv.org/abs/2610.21320
PDF
-
Making every stage of medical device document assessment with large language models deterministic
2026-10-15 05:45:20Ryosuke SuzukiAbstract
To show that a medical device complies with regulation, experts judge whether documents and records meet each requirement and record the judgment for others to verify. For each assessment item that experts define, large language models could prepare a preliminary assessment: a label such as Satisfied or Not satisfied, a reason and quoted evidence. We sent fourteen commercial deployments of such models (seven models from three providers, through six routes) the same request repeatedly, with the deterministic sampling settings each accepted; every deployment returned differing assessments for at least one request, and in twelve the label changed. Reducing reasoning where the interface allows it did not reduce the variation, and prompts of retrieved passages removed it only for the smallest models. The same open-weight model was reproducible or not depending only on how it was served. A pipeline we built to be deterministic at every stage -- text extraction and optical character recognition (OCR), chunking, embedding, indexing, retrieval and inference -- returned byte-identical output at each stage across reruns, interruptions, edits, machines of one graphics processing unit (GPU) model and days, for blood glucose meter and infusion pump documents. OCR, embedding and approximate retrieval varied when their own serving was not made invariant, and outputs changed with the GPU model and with the inference engine's scheduling settings, which therefore belong in the recorded configuration. End-to-end determinism is achievable and gives a testable basis for audit and change control.
Abstract (translated)
URL
https://arxiv.org/abs/2610.21300
PDF
-
Auditing Escalation Reliability in Healthcare LLM Agents: A Perturbation-Controlled Benchmark of Workflow-State Failures
2026-10-15 05:04:10Zihao Zheng (Washington University in St. Louis), Junyi Yao (Washington University in St. Louis), Baichuan Li (Southern Methodist University), Jiayu Long (Washington University in St. Louis)Abstract
Trustworthy biomedical AI requires reliable action selection when evidence is missing, stale, conflicting, or late, not only accurate answers on static inputs. We instantiate an episode-level evaluation protocol as 400 synthetic healthcare workflow episodes: 40 hand-authored semantic seeds across four task classes, two surface parameterizations, and five controlled variants, scored on ANSWER/ASK CLARIFICATION/ABSTAIN/ESCALATE under a pre-declared 5:1 missed-versus-unnecessary escalation cost. Reference actions are single-author study-policy labels, not clinician-adjudicated ground truth. Across five open-weight configurations (1.7B--9.7B), clean decision accuracy reaches 0.97, yet uncertainty-prompted models miss 32--97% of required escalations. The stronger 9.7B configuration narrows but does not remove the gap, falling from 0.97 clean accuracy to 0.78 under conflict. Verbalized confidence is inconsistent (ECE 0.10--0.54; failure AUROC 0.48--0.77), and a development-tuned confidence-only policy yields no utility gain. A transparent workflow-aware controller improves 5:1 utility for every model and reduces missed escalations to 0.12--0.24, at specificity 0.83--0.93. However, a 30-concept lexical stress test reduces each rule detector's recall from 1.00 on trigger phrases to 0.00 on unseen paraphrases, limiting the controller claim to this benchmark. Seed-clustered statistics, ablations, cross-task transfer, raw outputs, and manifests are packaged for audit. The results motivate explicit workflow-state evaluation while showing that apparently effective routing controls need their own generalization tests.
Abstract (translated)
URL
https://arxiv.org/abs/2610.21270
PDF
-
Really Need Recursion? Revisiting One-pass Registration with Multi-level Multi-space Contextual Guidance
2026-10-15 03:15:55Yue Lin, Shenghai Wu, Mingyuan MengAbstract
Deformable image registration is a fundamental challenge in medical image analysis. Recent deep learning-based methods typically adopt either one-pass architectures that complete registration in a single network iteration, or recursive architectures that refine registration through multiple passes. While recursive methods often improve accuracy, they incur substantial computational overhead and slower inference speed. In this paper, we challenge the necessity of recursion for high-performance deformable registration and propose M2CG, a one-pass registration network with Multi-level Multi-space Contextual Guidance. It performs multi-step progressive pyramid registration within a single iteration, guided at each step by contextual cues from three levels (previous, current, and subsequent steps) and four spaces (image space, image feature space, segmentation space, and segmentation feature space). Our M2CG supports both unsupervised (without any labels) and semi-supervised (with segmentation labels for partial training data) registration, and has achieved state-of-the-art performance across two well-benchmarked inter-/intra-patient medical registration tasks with eight public datasets. Our findings suggest that the current reliance on recursion may stem from the absence of sufficient contextual cues that comprehensively characterize the remaining distance to the target alignment.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22423
PDF
-
Quantifying privacy risks in foundation model representations of brain MRI
2026-10-14 17:03:42Imen Bajar, H\'eber Hwang Arcolezi, Antoine Boutet, Carole Frindel, Mohamed MaoucheAbstract
Pretrained representations are increasingly used to share and reuse information from medical images without exposing the original scans, but it is unclear how much patient-specific information these representations retain. We investigate this question for brain magnetic resonance imaging (MRI) by evaluating five frozen pretrained encoders spanning different architectures, pretraining objectives, and training domains on two longitudinal T1-weighted datasets, OASIS-2 and MIRIAD. Under an attacker model with access to released embeddings, the corresponding public encoder, and auxiliary MRI scans, we evaluate patient linkage and sensitive-attribute inference. Across the evaluated representations, we observe substantial subject-specific information: rank-1 patient linkage reaches up to 99\% on MIRIAD and exceeds 90\% on OASIS-2 for the highest-risk representation, despite the absence of an identification objective during encoder training. We further show that the released embeddings retain information about a demographic attribute, with recorded-sex prediction reaching balanced accuracy of 90.9\% on OASIS-2 and 95.5\% on MIRIAD. Overall, our results demonstrate that replacing raw brain MRI with learned embeddings does not, by itself, eliminate patient-linkage or attribute-inference risk, and is therefore insufficient on its own to establish anonymity.
Abstract (translated)
URL
https://arxiv.org/abs/2610.20494
PDF
-
Same Claim, Different Interpretations: How LLM Agents Resolve Ambiguity in Scientific Claims
2026-10-14 16:42:05Delip Rao, Ahmed Muharram, Chris Callison-BurchAbstract
Co-scientists and agentic scientific claim verifiers are increasingly used to assess claims across the sciences. Natural-language scientific claims, written for human experts, are often underspecified and so admit several defensible interpretations. Such verifiers usually pick one, often tacitly, and benchmarks score their verdicts against expert labels without recording the expert's interpretation or the verifier's in a comparable form. This leads to disagreements with expert labels that cannot be traced to the evidence or the interpretation, and to compute spent on unchecked interpretations. We introduce Claim Cards to make such interpretations explicit in the reasoning itself. A Claim Card records a verifier's interpretation as nine fields before evidence gathering, each marked stated, inferred, or default. To test whether an agent's interpretation can change its verdict, we ran a GPT-5.4 agent five times on each of 60 expert-labeled materials-science and biomedical claims with a card, without one, and with one card frozen across runs (900 runs). Over 40% of card fields were not marked stated, and a card added about a tenth to a run's GPT-5.4 tokens and time without measurably changing agreement with expert labels. The agent's own interpretations varied between runs, although within a claim this variation did not measurably track the verdict. A deliberate change to one field of the frozen card, however, making the quantifier universal or the threshold for support stricter, changed the verdict on about a third of the claims, against about one in seven expected by chance (p < 0.001 for each edit). On 24 materials-science claims run with computational tools, the stricter threshold also changed what the agent asked those tools to compute. We therefore argue that verdicts should be reported with their interpretations and that benchmarks could publish the interpretation behind each label.
Abstract (translated)
URL
https://arxiv.org/abs/2610.20457
PDF
-
Taking the PULSE of Medical Reasoning: Step-Level Retrieval Timing for Knowledge-Grounded Inference
2026-10-14 14:47:03Danyang Zhao, Sendong Zhao, Yanrui Du, Haochun Wang, Ming Ma, Wenhao Li, Jingbo Wang, Jiatong Liu, Bing QinAbstract
Reliable medical reasoning requires LLMs to dynamically access external biomedical evidence to improve factual trustworthiness and reasoning accuracy. However, indiscriminate retrieval can introduce noise and latency, raising a critical retrieval-timing problem: at each reasoning step, the system must decide whether the model's internal knowledge is reliable or external evidence is needed. We propose PUlSE, a lightweight probing method that uses the model's internal representations as intrinsic signals for step-level retrieval control. Our posterior trajectory analysis reveals that erroneous steps exhibit substantially higher inter-layer volatility than correct ones, suggesting unstable internal knowledge. Based on this finding, PUlSE trains a compact probe on inter-layer geometric features to detect factual errors upon step completion, and uses it as an adaptive trigger for retrieval and step regeneration. Across four medical benchmarks, PUlSE-timed retrieval matches exhaustive per-step retrieval in accuracy while reducing retrieval calls by 72%, and can transfer to an unseen legal domain, indicating potential domain-agnostic generality. Overall, our study provides an initial step toward shifting retrieval-trigger design from surface-level confidence estimates to internal activation dynamics for adaptive access.
Abstract (translated)
URL
https://arxiv.org/abs/2610.20292
PDF
-
Contrastive uncertainty learning for robust multi-structure segmentation in lower-pelvic MR
2026-10-14 14:23:02Bilal Sidiqi, Yipei Wang, Wen Yan, Shaheer U. Saeed, Daniel C. Alexander, Yipeng HuAbstract
Boundary delineation of anatomical structures in prostate MRI is often compromised by regional ambiguity. Automated segmentation methods relying on global overlap loss functions tend to be insensitive to surface deviations, leading to significant errors in ambiguous regions. Most existing uncertainty-aware methods treat uncertainty as a post-hoc quality indicator or prediction filter. We propose Contrastive Uncertainty Learning (CUL), a supervised framework that uses predictive uncertainty to directly guide representation learning. Using Monte Carlo Dropout, CUL identifies voxel-wise uncertainty and structures the embedding geometry accordingly: low-uncertainty, correctly predicted features form stable voxels while high-uncertainty or misclassified features are actively repelled, preventing feature collapse at difficult boundaries. We evaluate CUL on a 589-scan 3D prostate MRI multi-structure segmentation task (8 anatomical structures) using a ViT with UperNet decoder architecture. Average Dice ($83.38\%$) remains comparable to the baseline ($83.35\%$), while surface distance decreases significantly ($p<0.001$) across 6 of 8 anatomical structures. A hard-case analysis reveals a $22{\times}$ asymmetry between significantly improved and degraded cases ($18.8\%$ vs $0.9\%$). Ablation studies show that removing uncertainty-guided sample selection eliminates improvements for the transition zone, while replacing the contrastive objective with a boundary loss additionally removes bladder improvements, demonstrating that both uncertainty-guided sampling and the contrastive objective are essential for the observed gains. These findings demonstrate that the proposed uncertainty-guided representation learning is a useful mechanism for reducing the risk of extreme errors in safety-critical medical segmentation
Abstract (translated)
URL
https://arxiv.org/abs/2610.20251
PDF
-
MedSonor: Learning to Use Clinical Audio in Medical LLMs
2026-10-14 14:04:43Geyang Yu, Junying Chen, Benyou WangAbstract
Clinical audio conveys medical information that transcripts may not preserve, but using it in medical responses requires interpreting it alongside clinical requests and dialogue context. Existing medical audio supervision largely targets classification or recognition, while general audio-language models lack specialized medical knowledge and direct feedback for open-ended responses. We present MedSonor, a medical dialogue model integrating medical knowledge, spoken dialogue, and clinical audio understanding. We pair authentic medical audio with clinical requests and combine these audio-integrated dialogues with collected medical audio and synthesized spoken dialogues for supervised fine-tuning, followed by two stages of rubric-guided reinforcement learning. We introduce MedVoice-Dial and MedVoice?AID to evaluate responses within supplied multi-turn medical contexts and to requests involving clinical audio, respectively. Across seven medical benchmarks, MedSonor outperforms the evaluated general-purpose open-weight audio-language baselines and ranks highest in aggregate physician preference evaluation, while broadly retaining general speech capabilities.
Abstract (translated)
URL
https://arxiv.org/abs/2610.22405
PDF
-
Continual pretraining Prostate MRI ViT Encoder for task alignment
2026-10-14 13:57:23Bilal A. Sidiqi, Yipei Wang, Shaheer U. Saeed, Shiqi Huang, Daniel C. Alexander, Yipeng HuAbstract
Building foundation models for MRI has emerged as a promising direction for advancing medical image analysis and improving accessibility of task specific models. Existing approaches primarily focus on effective pretraining strategies and evaluating their performance across diverse downstream tasks. While these methods show performance gains, limited attention has been given to understanding task alignment between pretraining and downstream objectives. Unlike language models, where the two stages are naturally aligned, such alignment is less evident in vision based applications. Motivated by this gap, we explore continual pretraining. Specifically, continually pretraining a pretrained ViT encoder using Low-Rank Adaptation (LoRA) on a prostate segmentation task to enhance task alignment and downstream performance. We further perform layer-wise representational analysis using Centered Kernel Alignment (CKA) to assess representational shifts and their correlation with task outcomes. Our experiments show improved performance in 3 of 6 downstream tasks and comparable results in the others. Our Feature analysis reveals that, for segmentation tasks, continually pretrained encoder features exhibit greater similarity to those of the finetuned encoder than the original pretrained encoder, correlating with observed performance gains.
Abstract (translated)
URL
https://arxiv.org/abs/2610.20217
PDF
-
Dataset-Level Regularization for Joint-Embedding Self-Supervised Learning
2026-10-14 13:13:48Konstantinos Almpanakis, Anna KreshukAbstract
Self-supervised joint embedding learning aligns representations of augmented views while regularizing their distribution to prevent collapse. Regularization aims to shape the overall representation distribution, but most joint embedding methods restrict its computation to the active minibatch. We propose out-of-batch regularization, which maintains a dataset-level state in CPU memory and uses globally constructed per-instance targets during minibatch training. Evaluated on natural images and biomedical datasets, these anchors approximate the full-dataset regularization gradient more faithfully than conventional minibatch estimation at small batch sizes. Applications to contrastive (SimCLR) and non-contrastive (SIGReg) approaches show consistent improvement over the corresponding in-batch objectives for image classification, minority-class performance under controlled imbalance, instance retrieval, and continued self-supervised pretraining, with the largest gains at small batch sizes. These results suggest that dataset-level regularization can be decoupled from minibatch optimization, making batch size an optimization choice that does not affect the geometry of the regularizer.
Abstract (translated)
URL
https://arxiv.org/abs/2610.20158
PDF
-
Multimodal Medical Image Parsing with Task-Specific Low-Rank Adapters
2026-10-14 09:59:17Taha Sabir, Alberto Montero Molina, Thinh Quang Ly, Veronika Dmitrenko, Mohammad Yaqub, Tausifa Jan SaleemAbstract
Adapting a single vision-language model to medical image parsing requires training choices that reflect the demands of each task. We study task-specific adaptation with Qwen3.5-9B for FLARE 2026 Task 3, covering six task families and seven imaging modalities. Our system couples a shared, frozen 4-bit backbone with LoRA adapters selected from the input question. We customize adapter ranks and prompts, rebalance rare classification targets, and supplement regression training with dermatology classification examples. On the fixed internal holdout, the task-specific adapters achieve a dataset-macro balanced accuracy of 0.7791 for classification, a dataset-macro micro-F1 of 0.6286 for multi-label classification, a counting mean absolute error (MAE) of 11.73, a dataset-macro detection F1 of 0.8914, a dataset-macro regression MAE of 9.78, and a report-generation CRIMSON score of 0.8204. On the organizer's 1,703-example hidden-validation set, the submitted system achieves 0.8496 classification balanced accuracy, 0.5338 sample-averaged multi-label F1, and 14.5950 regression MAE. The final 30.1 GB container averages 3.1 seconds per case, including model loading, and produces identical outputs across repeated verification runs. The benefits vary across tasks, supporting the use of task-specific adapter configurations and image budgets within a shared medical vision-language model. Code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.19901
PDF
-
PRISMATIC : Prescription Risk Inspection System for Multi-Agent Treatment Interaction in Clinical Decision-Making
Abstract
Medical prescribing errors for multimorbid patients, who require complex drug regimens, are a major patient safety concern. These errors often stem from fragmented decision-making across medical specialties, a critical coordination failure that leads to adverse drug events and rehospitalizations. To address this coordination challenge, we introduce PRISMATIC, a multi-agent architecture where specialized agents, each representing a workflow role, collaboratively generate and refine treatment plans. Operating on a blackboard architecture, the system first structures patient information from clinical notes, then iteratively develops safe, rational, and interpretable drug regimens. The agents' collaboration is governed by two novel adaptive coordination mechanisms: (1) Dynamic Self-Updating Weight Adjustment (DSWA), a long-term mechanism that adjusts the influence weights of agents based on historical case outcomes, enabling the system to adapt and trust more reliable specialists over time. (2) Discrepancy Feedback Precise Calibration (DFPC), a short-term mechanism where agents provide iterative feedback to resolve disagreements and eliminate prescription conflicts for the current patient. Experiments on patient records from the MIMIC-IV database show that PRISMATIC significantly reduces drug conflicts, improves prescription coverage, and enhances treatment consistency compared to strong LLM baselines. Our work demonstrates that structuring clinical co-treatment as a multi-agent coordination problem offers a promising path toward safer and more coherent multidisease management.
Abstract (translated)
URL
https://arxiv.org/abs/2610.19762
PDF
-
Beyond Reachability: A Relation-Semantic Reasoning Framework for Biomedical Multi-hop QA
2026-10-14 08:09:02Ziyang Ling, Ziyi Chen, Ronald X. Xu, Mingzhai SunAbstract
Biomedical multi-hop reasoning over knowledge graphs requires identifying a semantically valid relation sequence that supports the answer. Existing methods can retrieve correct answers, yet often rank them poorly within a large pool of seemingly relevant candidates. Even when a correct answer is retrieved, its associated evidence and reasoning path may fail to support the relation composition required by the question. We term this disconnect between answer reachability and evidential validity the reachable-but-invalid problem. We present ReSemReason, which shares a question-induced distribution over typed, directed relation schemas between candidate discovery and evidence selection. Schema compatibility guides graph search; ReSemAttn then uses relation-conditioned key/value projections and an uncertainty-weighted path prior to attend to candidate-specific evidence. A separate absolute-support score complements normalized attention when all retrieved paths are weak. The reported experiments on BioHopR, PrimeKGQA, and MedReason show improvements in answer ranking and answer-set quality. Path-discrimination analyses examine the distinction between endpoint correctness and evidential support. We also specify matched controls for search, attention, and schema uncertainty.
Abstract (translated)
URL
https://arxiv.org/abs/2610.19754
PDF
-
SMI: Ontology-Grounded Semantic IDs for Hierarchy-Aware Medical Concept Representation
2026-10-13 16:30:01Ziyang Song, Lia Shen, Yixuan Li, Qincheng Lu, Weihao Li, Haoran Xu, He Zhu, Xinyu Wang, Yue LiAbstract
Biomedical language models (LMs) provide strong representations for clinical tasks, but they often fail to capture the hierarchical structure among medical concepts. This limitation is especially pronounced for rare and fine-grained clinical codes in long-tailed vocabularies, where hierarchy-agnostic LM representations become less informative as concept frequency decreases. We introduce Semantic Medical Identifier (SMI), a training-free framework for constructing ontology-grounded Semantic IDs and hierarchy-aware medical concept representations. For each diagnosis, procedure, or medication concept, SMI defines its Semantic ID as the coarse-to-fine path through an expert-defined medical ontology. SMI then aggregates frozen biomedical LM embeddings across the corresponding ontology levels to form hierarchy-aware medical concept representations that preserve coarse-to-fine clinical semantics. The base SMI uses uniform aggregation across ontology levels, while a level-weighted variant selects global level weights on validation data. Our theoretical analysis shows that SMI yields more informative medical concept representations than hierarchy-agnostic LM embeddings, particularly for rare and fine-grained clinical concepts. Experiments show that SMI captures meaningful hierarchical structure and improves clinical relation prediction. On MIMIC-IV, SMI improves mortality prediction, readmission prediction, drug recommendation, and rare drug prediction. On eICU, SMI yields more robust patient representations under cross-hospital distribution shift.
Abstract (translated)
URL
https://arxiv.org/abs/2610.18672
PDF
-
Geometric Calibration of Supervised Fine-Tuning
2026-10-13 16:07:24Jianzhu Bao, Jia-Chen Gu, Qingyuan Liu, Haozhen Zhang, Kai-Wei Chang, Nanyun Peng, Wenya WangAbstract
Recent studies investigate the weight geometry through its spectral structure, which characterizes how the weights transform different input directions. Within this view, supervised fine-tuning (SFT) is shown to induce dense updates that perturb the spectral geometry of pre-SFT weights along their principal directions, potentially disrupting the model's prior knowledge. These observations motivate using the geometry of pre-SFT weights as a reference for refining learned SFT updates. We study this perspective as Geometric Calibration and develop SPACE Subspace Projection for Adapted Checkpoint Enhancement), a lightweight post-hoc method that selectively reverts SFT update components within the principal subspace of the pre-SFT weights while preserving those in its orthogonal complement. SPACE is a data-free, training-free method that requires only the pre- and post-SFT checkpoints. We evaluate SPACE on Qwen2.5, Qwen3, Qwen3.5, Llama-3.1, and Olmo-3 models ranging from 2B to 14B, across mathematical, coding, and medical SFT, as well as several SFT variants. Overall, SPACE mitigates out-of-domain (OOD) capability regression while further improving in-domain (ID) performance over the uncalibrated SFT checkpoints. For math SFT, SPACE yields average ID/OOD gains of 2.85/2.26 points on Qwen2.5-7B, and 2.41/1.52 points on Qwen3-14B-Base, respectively. These results highlight how the geometric relationship between SFT updates and pre-SFT weights shapes the adapted model's capabilities and, more broadly, motivate geometry-aware parameter updates for continual model improvement.
Abstract (translated)
URL
https://arxiv.org/abs/2610.18634
PDF
-
Chain of Thought Monitoring Can Fail Even When CoT Is Necessary
2026-10-13 10:22:01Lennart B\"urger, Piotr Komorowski, Moritz Weckbecker, Evgenii Kortukov, Aakriti Jain, Sebastian Lapuschkin, Wojciech SamekAbstract
Chain of Thought (CoT) monitoring is used by frontier labs to detect misbehaviour of AI systems. Yet, its reliability remains poorly understood. We attribute this to the lack of a successful organizing principle that leads to generalizable conclusions about monitoring reliability in different regimes. Recent work attributes monitoring failures to the CoT not being necessary for the answer, and argues that CoT becomes monitorable when it is necessary. We show that this binary necessary/not-necessary organizing principle is too coarse by demonstrating that monitoring failures persist even if the CoT is necessary for the answer. In response, we introduce a more fine-grained organizing principle based on a hierarchy of four CoT necessity levels. Building on this hierarchy, we develop a framework for measuring CoT monitorability at the three lowest levels via intervention experiments. We demonstrate monitoring failures at all three of these levels on realistic, hard prediction tasks. On a difficult medical-diagnosis task, patient sex and a clinical attribute influence the answer in at least 19% and 24% of correct responses with a necessary CoT, respectively, yet their influence is almost never articulated. These results show that even when CoT is necessary, it is not guaranteed to be monitorable.
Abstract (translated)
URL
https://arxiv.org/abs/2610.18105
PDF
-
inLoRA-MINT: Internal Representations for Auditing Training Data in Domain-adapted LLMs
2026-10-13 09:40:58Gonzalo Mancera, Julian Fierrez, Aythami Morales, Ruben Tolosana, Daniel DeAlcala, Francisco JuradoAbstract
This work investigates the application of the Membership Inference Test (MINT) to large language models (LLMs) fine-tuned with LoRA, aiming to determine whether specific samples were used during training and address concerns about data privacy, leakage, and transparency. Building on our previous LoRA-MINT framework, which relies exclusively on perplexity under a strict black-box setting, we propose an approach based on intermediate model representations, including activations and parameter-dependent signals. These representations enable a MINT classifier to distinguish training from non-training samples without requiring full access to model parameters. We evaluated the method on four LLMs (Qwen, Phi-4, DeepSeek-V3, and LLaMA 3.2-3B) fine-tuned on three datasets covering medical, mathematical, and academic reasoning tasks. Results show consistent performance across architectures and domains, achieving scores between 95% and 98%, with robustness to variations in model configurations and data distributions. The proposed approach enables efficient auditing of sensitive training data and supports data provenance and compliance with emerging privacy and AI governance regulations. Although our experiments focus on LoRA-based fine-tuning, the methodology can be extended to other LLM adaptation techniques and, more broadly, to other AI models.
Abstract (translated)
URL
https://arxiv.org/abs/2610.18051
PDF
-
Seeing Through the Forgery Is Not Enough: Regressive Sycophancy under Fabricated Evidence in Medical Vision-Language Models
2026-10-13 08:55:06Sadif Ahmed, Soumya Swagata Biswas, Rifat Shahriyar, Chowdhury Mohammad Rakin HaiderAbstract
A medical vision-language model may analyze a given CT scan and correctly detect both kidneys in view. Continuing this conversation, if a user provides a prescription with an opposite direction, medical VLMs often sycophantically choose to realign their diagnosis. In this work, we study regressive sycophancy under fabricated evidence, where a model abandons a correct answer when pushback comes with a fake document. We test four open VLMs on 400 VQA-RAD yes/no questions in 19,200 multi-turn conversations, with and without a fabricated prescription. Without a disclosure that documents are synthetic, the models abandon 80.1% of 6,488 correct answers. Notably, under one script each, the same prescription raises the flip rate of Qwen2.5-VL-7B from 45.9% to 89.6% and lowers that of Qwen3-VL-8B from 77.7% to 50.9%. A model rejected the prescription while still preserving the correct answer in 915 of 3,232 conversations with the prescription (28.3%), and 40% of these flipped in later turns. System-prompt sentences that disclose the documents as synthetic lower one model's flip rate from 77.8% to 47.3%. Thus, we conclude that awareness about a forgery does not prevent a model from adopting its answer.
Abstract (translated)
URL
https://arxiv.org/abs/2610.17989
PDF
