Search: “label-leakage”

Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI

Study of label leakage and anatomical grounding in multimodal MRI models for Alzheimer's staging shows cognitive-score fusion accuracy of 87.3% is leakage-driven.

The authors train a ResNet18 slice-based encoder with a one-layer Transformer on 1,075 ADNI-1 T1 MRI scans, using FastSurfer segmentations and YOLOv8 localization (mAP_50 above 0.96) as anatomical reference. Grad-CAM shows the image-only classifier often attends to skull and background rather than disease-relevant structures. A CLIP-style image-tabular contrastive framework organized along a label-leakage spectrum yields 87.3% three-way accuracy with cognitive scores versus 73.0% with regional volumes, and cropping to the medial temporal lobe raises image-only accuracy from 58.7% to 65.1%. Results come from single runs on a small balanced test set with reported confidence intervals.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 17:13 UTC · 2d agoAI research1

VLC Media Player Flaws Let Attackers Corrupt Memory and Leak Sensitive Data

VLC 3.0.0-3.0.23 has two flaws: a heap out-of-bounds write via malicious PNGs (CVE-2026-56711) and an out-of-bounds read via hostile RealRTSP servers.

CVE-2026-56711 is a heap out-of-bounds write (CVSS v4 8.6) caused by an integer overflow in VLC's AllocatePicture routine when processing PNG images with exceptionally large IHDR width and height values, allowing writes past the allocated buffer; it maps to CWE-190 and CWE-787 and was credited to Fabian Wahle of Hap Security. CVE-2026-73324 is a medium-severity out-of-bounds read (CVSS v4 6.9) in VLC's RealRTSP handling, where RtspReadLine copies response lines longer than 4,096 bytes into a fixed buffer without null termination, potentially leaking heap data back to a hostile RTSP server via the Session header. Both bugs affect VLC 3.0.0 through 3.0.23, and updated builds had not yet been released at the time of disclosure.

GBHackers·updated Sep 12, 14:31 UTC · 4d ago·first Sep 11, 12:03 UTC · 5d agoVulnerability 2 sourcesCVE-2026-56711CVE-2026-73324

ZDI-26-670: Adobe Acrobat Pro DC Doc Object Out-Of-Bounds Read Information Disclosure Vulnerability

ZDI published advisory ZDI-26-670 for an out-of-bounds read information disclosure flaw (CVE-2026-81991) in Adobe Acrobat Pro DC.

The Zero Day Initiative disclosed ZDI-26-670, an out-of-bounds read in the Doc object of Adobe Acrobat Pro DC. A remote attacker could disclose sensitive information from affected installations if the user opens a malicious file or page. ZDI rated the issue 3.3 on the CVSS scale and assigned CVE-2026-81991.

ZDI Published Advisories·Sep 10, 05:00 UTC · 7d agoAdvisory 19 sourcesCVE-2026-81991

Target leakage, not model class, explains reported accuracy in survey-based cardiovascular screening: a leakage-tiered audit of glass-box and tabular foundation models

Audits of 10 classifiers on BRFSS show target leakage, not model class, drives the reported 0.89 AUROC in survey-based cardiovascular screening.

The study benchmarks ten model classes, including glass-box and tabular foundation models, for prevalent myocardial infarction on 442,067 respondents of the 2022 BRFSS across five feature tiers of decreasing leakage risk. Removing two post-diagnostic features costs every model 0.049-0.051 AUROC and collapses performance into a 0.0045-wide band, and the explainable boosting machine matches all alternatives within 0.005 while scoring roughly 104x faster than the strongest foundation model. Frozen models transport within 0.002 AUROC to 2023 data; the authors conclude evaluation practice and feature sets, not model capacity, are the binding constraint.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:29 UTC · 6d agoAI research1

General Quantification of Covariate and Concept Shifts

Paper proposes γ*-concept shifts via entropic optimal transport, deriving estimable generalization bounds unifying covariate and concept shift under distribution shift.

The authors show existing definitions of concept shift break when source and target supports mismatch and propose γ*-concept shifts grounded in entropic optimal transport. They derive a general error bound covering broad loss functions, label spaces and stochastic labeling, plus estimators with concentration guarantees. The resulting DataShifts algorithm quantifies distribution shifts and estimates the error bound in most applications, addressing learning bounds that were previously non-estimable from samples.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:57 UTC · 6d agoAI research1

Client-Side Probing of Deleted Ridge Statistics in Federated Unlearning

Researchers show malicious federated learning clients can probe broadcast classifiers to recover deleted samples, exposing exact label leakage on MNIST and CIFAR-10.

The paper shows that federated unlearning systems broadcasting updated linear classifiers leak compact additive training summaries to clients. A malicious client can submit known changes, identify server states from returned classifiers, and compare states around an isolated deletion to expose the deleted sample, class, or client summary, potentially enabling reinsertion. On MNIST and CIFAR-10, high-precision broadcasts allowed exact label recovery for every tested deletion, while lower precision sharply reduced fine-grained recovery.

arXiv cs.CR·Sep 3, 20:59 UTC · 13d agoResearch

[webapps] Duplicati 2.2.0.3 - JWT Signing Key Leak

A public exploit exposes JWT signing key leakage in backup software Duplicati 2.2.0.3, risking session forgery.

Exploit-DB published exploit #52646 for Duplicati 2.2.0.3, a backup application. The issue leaks the JWT signing key, which could let an attacker forge authentication tokens. The disclosure does not report any exploitation in the wild.

Exploit-DB·Aug 17, 00:00 UTC · Aug 17, 2026Exploit / PoC

The Illusion of Local Privacy: Confidentiality Boundary Failures in Consumer LLM Serving Systems

Researchers show local LLM serving systems leak prompts via memory residue, plaintext persistence, a llama.cpp tenant-isolation flaw, and timing oracles.

A study of consumer local-LLM serving systems identifies four boundaries where prompt confidentiality fails: model loading, runtime memory, wrapper persistence, and the serving interface. Using the LLAnalyzer framework across four open-weight model families and two deployment platforms, the authors recover plaintext prompts from allocator-managed memory after inference and show wrappers extend prompt lifetime. They also uncover a previously undocumented llama.cpp authorization flaw letting one authenticated client restore another tenant's saved conversation state, succeeding in 200/200 trials, plus a remote timing oracle via shared prompt-prefix caching that works over WAN.

arXiv cs.CR·Sep 16, 11:53 UTC · 17h agoAI safety & security

Privacy Failure in Split-LLM Training, The Returned Gradient Nullifies the Decoys

Researchers show split-LLM training leaks privacy via zero-valued gradients on decoy rows, exposing which activations are real despite passing forward-channel checks.

A systems-security case study of a two-node split-LLM training setup found that the returned output gradient from an Untrusted Cloud Node is exactly zero for decoy rows, revealing which rows are real. Across nine seeds, zero patterns identified real rows in 4,096 of 4,096 frames per run, and an attack on frame contents recovered 0.65 to 1.50 percentage points of extra tokens over a baseline. Both datasets passed forward-channel privacy and quality checks but failed once the returned gradient was included. Row-wise gradient clipping and noise closed the leak for roughly 0.01 nats of held-out cross-entropy, though five unmeasured attack classes remain.

Hugging Face daily papers·Sep 2, 20:00 UTC · 14d agoAI safety & security

Subgroup Membership Inference Audits of Differentially Private Synthetic Text

Audits of 32 differentially private synthetic-text releases show subgroup membership leakage is concentrated in few records and systematically underestimated by average-case attacks.

The paper defines a subgroup-targeted membership inference game in which the target pool is an explicit parameter, to audit residual leakage in differentially private synthetic text releases. The audit instantiates 32 proxies across four datasets, three generators (DP-SGD fine-tuning, API-based prompting, and activation steering), and five privacy budgets. DP substantially reduces average leakage at every budget, but remaining leakage is concentrated: roughly a tenth of records carries about 40% of it, and the noise removes more measured leakage from random records than from high-risk ones. Which records leak depends on the release mechanism, so record-level risk cannot be assessed independently of the release.

arXiv cs.CR·Sep 9, 07:56 UTC · 7d agoResearch

ZDI-26-613: (0Day) pdfforge PDF Architect PDF File Parsing Memory Corruption Remote Code Execution Vulnerability

ZDI published ZDI-26-613, an unpatched memory corruption flaw in pdfforge PDF Architect PDF parsing enabling remote code execution (CVSS 7.8).

The Zero Day Initiative disclosed ZDI-26-613, a memory corruption vulnerability in pdfforge PDF Architect's PDF file parsing that allows remote code execution on affected installations. User interaction is required, meaning the target must visit a malicious page or open a malicious file. ZDI assigned a CVSS score of 7.8 and classifies the issue as a 0day.

ZDI Published Advisories·Aug 31, 05:00 UTC · 17d agoAdvisory

Closing the Blind Spot: Securing Personal Repositories in the Software Supply Chain

Wiz highlights personal developer repositories as a supply chain blind spot leaking corporate secrets, offering correlation-based risk validation and remediation.

Wiz argues that developers' personal code repositories are a blind spot in software supply chain security where corporate secrets quietly escape. The company describes an approach that correlates personal repositories to specific developers, validates the actual risk, and drives remediation. No specific incident or vulnerability is disclosed in the announcement.

Wiz Blog·Aug 13, 13:48 UTC · Aug 13, 2026Tools2

ZDI-26-550: OriginLab OriginPro OGW File Parsing Memory Corruption Remote Code Execution Vulnerability

ZDI discloses a memory corruption flaw in OriginLab OriginPro OGW file parsing enabling remote code execution via malicious files (CVE-2026-18291, CVSS 7.8).

ZDI advisory ZDI-26-550 describes a memory corruption vulnerability in OriginLab OriginPro OGW file parsing that allows remote attackers to execute arbitrary code. User interaction is required: the target must visit a malicious page or open a malicious file. The flaw has a CVSS rating of 7.8 and is assigned CVE-2026-18291.

ZDI Published Advisories·Aug 11, 05:00 UTC · Aug 11, 2026VulnerabilityCVE-2026-18291

ZDI-26-630: NI LabVIEW VI File Parsing Integer Overflow Information Disclosure Vulnerability

ZDI disclosed CVE-2026-18445, an integer overflow in NI LabVIEW VI file parsing that can disclose sensitive information, rated CVSS 3.3.

The Zero Day Initiative published advisory ZDI-26-630 describing an integer overflow vulnerability in NI LabVIEW's parsing of VI files. Exploitation can disclose sensitive information and requires user interaction, such as visiting a malicious page or opening a malicious file. ZDI assigned the flaw a CVSS rating of 3.3.

ZDI Published Advisories·Sep 9, 05:00 UTC · 8d agoAdvisory 2 sourcesCVE-2026-184451

Nearly 750k had financial info, SSNs leaked in South Carolina loan company breach

Heights Finance breach of a third-party cloud platform exposed SSNs and banking data of 734,828 loan customers across 11 states.

Attackers breached a third-party cloud platform used by Heights Finance in May, exposing data on 734,828 customers, according to the company's filing with Texas regulators. Stolen data includes contact details, bank account and routing numbers, Social Security numbers, tax IDs and driver's license numbers. The breach, discovered on May 7, was limited to the cloud platform and did not affect loan management systems. No group has claimed the attack and dark web monitoring has found no evidence of the data being leaked.

The Record·Aug 17, 20:14 UTC · Aug 17, 2026Data breach

ZDI-26-542: Microsoft Windows UMPDDrvBitBlt Improper Object Management Local Privilege Escalation Vulnerability

ZDI discloses CVE-2026-62712, a CVSS 7.8 Windows UMPDDrvBitBlt improper object management flaw allowing local attackers to escalate privileges.

ZDI advisory ZDI-26-542 describes improper object management in Microsoft Windows' UMPDDrvBitBlt function, tracked as CVE-2026-62712 with a CVSS score of 7.8. The flaw allows local attackers to escalate privileges on affected Windows installations. Exploitation requires first obtaining the ability to execute low-privileged code on the target system.

ZDI Published Advisories·Aug 11, 05:00 UTC · Aug 11, 2026VulnerabilityCVE-2026-627122

Label-Guided Knowledge Distillation for 3D-CNNs in Action Recognition

LGKD uses ground-truth labels to guide feature distillation for 3D-CNNs, combining sample-wise and class-wise distillation for action recognition.

The paper proposes Label-Guided Knowledge Distillation (LGKD) for 3D-CNNs, noting that most video feature distillation methods are simple adaptations of image techniques that neglect temporal-dimension differences. LGKD combines sample-wise distillation, which uses label information and the teacher's probability distribution to guide features impacting temporal accuracy, with class-wise distillation employing a prototype network to capture relational knowledge among same-category samples. Experiments on the UCF101 and HMDB51 action recognition benchmarks achieve competitive results.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 16:21 UTC · 5d agoAI research

Reference-Based Bias Detection in LLMs via Relative Representations of Hidden States

Researchers propose auditing LLM bias via relative hidden-state representations, detecting bias increases with 3-50x less compute than output-level benchmarks.

The paper introduces a reference-based bias auditing method that compares hidden-state representations across model variants, such as before and after fine-tuning, by encoding sentences relative to a fixed anchor set. The resulting Representational Bias Shift (Delta-B) correlates with output-level bias change in 15 of 18 tested settings, reaching |r| = 0.84 under full fine-tuning across WildGuardMix, DecodingTrust, and ToxiGen benchmarks. Thresholding Delta-B detects checkpoints whose bias increased with ROC AUC between 0.65 and 0.99 and beats a SEAT-based baseline, while auditing a model in about three minutes with 3-50x less compute.

Hugging Face daily papers·Sep 8, 20:00 UTC · 8d agoAI research1

[Control Systems] National Instruments security advisory (AV26-856)

Canada's Cyber Centre relayed National Instruments advisories for memory corruption, out-of-bounds read, and out-of-bounds write flaws in LabVIEW versions.

The Canadian Centre for Cyber Security published control systems advisory AV26-856 covering National Instruments LabVIEW. Affected versions include releases before 23.0.0, 23.3.10, 24.3.7, 25.3.5, and 26.3.1. The flaws include memory corruption, an integer conversion out-of-bounds read, and an integer overflow out-of-bounds write. Users and administrators are urged to review the links and apply NI security updates.

Canadian Centre for Cyber Security·Aug 28, 12:16 UTC · 19d agoAdvisory

ZDI-26-621: Microsoft Windows UMPDDrvRealizeBrush Improper Object Management Local Privilege Escalation Vulnerability

ZDI disclosed CVE-2026-62712, a CVSS 7.8 improper object management flaw in Windows UMPDDrvRealizeBrush enabling local privilege escalation after low-privileged code execution.

The Zero Day Initiative published ZDI-26-621 covering a local privilege escalation vulnerability in Microsoft Windows' UMPDDrvRealizeBrush component, caused by improper object management. An attacker must already be able to execute low-privileged code on the target system before exploiting the flaw. ZDI assigned the vulnerability a CVSS score of 7.8 and the CVE identifier CVE-2026-62712.

ZDI Published Advisories·Sep 8, 05:00 UTC · 9d agoAdvisory 4 sourcesCVE-2026-627121

New cPanel Critical Flaw Could Let Hosting Customers Run SQL as Database Root

The Hacker News·Aug 4, 10:36 UTC · Aug 4, 2026VulnerabilityCVE-2026-58048CVE-2026-58047

GDCM <= 3.2.7: six memory-safety and denial-of-service vulnerabilities, no CVE

Six memory-safety and denial-of-service flaws disclosed in the GDCM DICOM parsing library, affecting versions through 3.2.7.

Researcher Abhinav Agarwal disclosed six vulnerabilities in GDCM (Grassroots DICOM), an open-source C++ library for parsing and processing DICOM files. All six were confirmed against GDCM 3.2.6 using AddressSanitizer and UndefinedBehaviorSanitizer, and source review found the vulnerable patterns through version 3.2.7 and the upstream master snapshot. Potential impacts include heap corruption, process-memory disclosure, stack exhaustion, and process termination in applications parsing untrusted DICOM files. No CVE identifiers have been assigned at the time of disclosure.

oss-security·Sep 10, 03:35 UTC · 7d agoVulnerability

Normal Alignment: Improved Cryptanalytic Sign Recovery on Hard-Label Networks

Researchers propose Normal Alignment, improving cryptanalytic sign recovery for hard-label neural networks and enabling polynomial-time full model extraction.

The paper improves on Carlini et al.'s EUROCRYPT 2025 cryptanalytic extraction of hard-label (S1) DNNs, whose Future Toggle sign-recovery method offered only marginal advantage over random guessing and triggered exponential-time enumeration on errors. Normal Alignment infers neuron signs via expected length differences between projected normals of adjacent decision facets at dual points, delivering higher voting accuracy and low-confidence errors. Combined with the SOE extension, it achieves exact polynomial-time full sign recovery: CIFAR-10 (192-64x8-10) and MNIST (64-96x3-32-10) models are fully recovered where the prior method required 2^52 or 2^82 sign guesses.

arXiv cs.CR·Sep 16, 14:42 UTC · 14h agoResearch

ATM Flaws Reveal Key Weaknesses in the Software Supply Chain

A researcher disclosed nine vulnerabilities in ATM encryption and authentication software, highlighting weaknesses across the software supply chain.

WIRED reports that a security researcher found nine vulnerabilities in software used for ATM encryption and authentication. The disclosure matters beyond cash machines because the affected components illustrate broader weaknesses in the software supply chain. The article does not report active exploitation of the flaws.

WIRED · Security·Aug 31, 10:00 UTC · 16d agoVulnerability

Re: Vulnerabilities fixed in libxml2-2.15.4

libxml2 2.15.4 patches two flaws including a heap buffer overflow in xmlDictAddQString tracked as CVE-2026-86137 and CVE-2026-86138.

libxml2 releases before 2.15.4 are affected by an integer overflow in xmlDictAddQString in dict.c that leads to a heap-based buffer overflow, tracked as CVE-2026-86137 and CVE-2026-86138. The oss-security post from Debian's Salvatore Bonaccorso flags the fixed release for downstream tracking. No exploitation is mentioned in the disclosure.

oss-security·Sep 5, 12:16 UTC · 11d agoVulnerabilityCVE-2026-86137CVE-2026-86138

ZDI-26-605: Microsoft Windows Localized Filenames Improper Input Validation NTLM Response Information Disclosure Vulnerability

ZDI advisory ZDI-26-605 details an improper input validation flaw (CVE-2026-50508, CVSS 3.3) in Microsoft Windows localized filenames that leaks NTLM responses.

The Zero Day Initiative released advisory ZDI-26-605 describing improper input validation in Microsoft Windows handling of localized filenames. Remote attackers can disclose NTLM authentication responses if the target opens a malicious file or visits a crafted page. ZDI rated the issue CVSS 3.3 and assigned CVE-2026-50508. Leaked NTLM responses could enable offline credential cracking.

ZDI Published Advisories·Aug 24, 05:00 UTC · 24d agoAdvisoryCVE-2026-505081

Mythos Vulnerability Firehose Hits a Human Bottleneck

Analysis of Project Glasswing findings shows only a fraction of discovered vulnerabilities have reached disclosure and even fewer are fixed.

Dark Reading reports that an analysis of Project Glasswing findings shows only a fraction of the vulnerabilities discovered by the program have reached disclosure, and an even smaller number have been fixed. The article examines how the volume of findings from the discovery program is bottlenecked by limited human triage and remediation capacity. The piece highlights growing tension between high-volume vulnerability discovery and the industry's ability to process, disclose and patch reports.

Dark Reading·Sep 9, 21:19 UTC · 7d agoResearch

Rare Not Random Using Token Efficiency for Secrets Scanning

Researcher proposes token efficiency (string length divided by BPE token count) as a better post-regex filter than entropy for secrets scanning, validated on CredData.

The post explores whether Byte-Pair Encoding tokenization can replace Shannon entropy as the primary filter for candidate secrets captured by regex in tools like Gitleaks. It defines 'token efficiency' as string length divided by token count under the cl100k_base tokenizer; secret-like strings such as GitHub tokens tokenize into many small tokens and score low, while natural text scores high. Evaluating labeled secrets from the CredData dataset shows a usable separation, with roughly 2.5 suggested as a minimum cutoff versus Gitleaks' 3.5 entropy threshold. The technique is positioned as a post-regex filtering step rather than a standalone detector.

Lobsters · security·Sep 12, 06:42 UTC · 4d agoResearch