Aggregator

A vulnerability categorized as critical has been discovered in EFM iptime A6004MX 14.18.2. Affected is the function commit_vpncli_file_upload of the file /cgi/timepro.cgi. The manipulation results in unrestricted upload. This vulnerability was named CVE-2026-2550. The attack may be performed from remote. In addition, an exploit is available. Applying restrictive firewalling is recommended. The vendor was contacted early about this disclosure but did not respond in any way.

Submit #749986 / VDB-346159

A vulnerability was found in zhanghuanhao LibrarySystem 图书馆管理系统 up to 1.1.1. It has been rated as critical. This impacts an unknown function of the file BookController.java. The manipulation leads to improper access controls. This vulnerability is uniquely identified as CVE-2026-2549. The attack is possible to be carried out remotely. Moreover, an exploit is present. The project was informed of the problem early through an issue report but has not responded yet.

A vulnerability was found in WAYOS FBM-220G 24.10.19. It has been declared as critical. This affects the function sub_40F820 of the file rc. Executing a manipulation of the argument upnp_waniface/upnp_ssdp_interval/upnp_max_age can lead to command injection. This vulnerability is handled as CVE-2026-2548. The attack can be executed remotely. Additionally, an exploit exists. The vendor was contacted early about this disclosure but did not respond in any way.

Submit #749873 / VDB-346158

A vulnerability was found in LigeroSmart up to 6.1.26. It has been classified as problematic. The impacted element is the function AgentDashboard of the file /otrs/index.pl. Performing a manipulation of the argument Subaction results in cross site scripting. This vulnerability is known as CVE-2026-2547. Remote exploitation of the attack is possible. Furthermore, an exploit is available. The project was informed of the problem early through an issue report but has not responded yet.

A vulnerability was found in LigeroSmart up to 6.1.26 and classified as problematic. The affected element is an unknown function of the file /otrs/index.pl. Such manipulation of the argument SortBy leads to cross site scripting. This vulnerability is traded as CVE-2026-2546. The attack may be launched remotely. Furthermore, there is an exploit available. The project was informed of the problem early through an issue report but has not responded yet.

A vulnerability has been found in LigeroSmart up to 6.1.26 and classified as problematic. Impacted is an unknown function of the file /otrs/index.pl?Action=AgentTicketSearch. This manipulation of the argument Profile causes cross site scripting. This vulnerability appears as CVE-2026-2545. The attack may be initiated remotely. In addition, an exploit is available. The project was informed of the problem early through an issue report but has not responded yet.

Submit #749802 / VDB-346157

Больше никаких пуль и гильз — только беззвучный луч и запах горелой электроники.

Submit #749788 / VDB-346156

Submit #749784 / VDB-346155

Submit #749758 / VDB-346154

Session 12C: Membership Inference

Authors, Creators & Presenters:

PAPER
Yuefeng Peng (University of Massachusetts Amherst), Ali Naseh (University of Massachusetts Amherst), Amir Houmansadr (University of Massachusetts Amherst)

Deep learning models, while achieving remarkable performances across various tasks, are vulnerable to membership inference attacks (MIAs), wherein adversaries identify if a specific data point was part of the model's training set. This susceptibility raises substantial privacy concerns, especially when models are trained on sensitive datasets. Although various defenses have been proposed, there is still substantial room for improvement in the privacy-utility trade-off. In this work, we introduce a novel defense framework against MIAs by leveraging generative models. The key intuition of our defense is to *remove the differences between member and non-member inputs*, which is exploited by MIAs, by re-generating input samples before feeding them to the target model. Therefore, our defense, called Diffence, works *pre inference*, which is unlike prior defenses that are either training-time (modify the model) or post-inference time (modify the model's output). A unique feature of Diffence is that it works on input samples only, without modifying the training or inference phase of the target model. Therefore, it can be cascaded with other defense mechanisms as we demonstrate through experiments. Diffence is specifically designed to preserve the model's prediction labels for each sample, thereby not affecting accuracy. Furthermore, we have empirically demonstrated that it does not reduce the usefulness of the confidence vectors. Through extensive experimentation, we show that Diffence can serve as a robust plug-n-play defense mechanism, enhancing membership privacy without compromising model utility--both in terms of accuracy and the usefulness of confidence vectors--across standard and defended settings. For instance, Diffence reduces MIA attack accuracy against an undefended model by 15.8% and attack AUC by 14.0% on average across three datasets, all without impacting model utility. By integrating Diffence with prior defenses, we can achieve new state-of-the-art performances in the privacy-utility trade-off. For example, when combined with the state-of-the-art SELENA defense it reduces attack accuracy by 9.3%, and attack AUC by 10.0%. Diffence achieves this by imposing a negligible computation overhead, adding only 57ms to the inference time per sample processed on average.

ABOUT NDSS
The Network and Distributed System Security Symposium (NDSS) fosters information exchange among researchers and practitioners of network and distributed system security. The target audience includes those interested in practical aspects of network and distributed system security, with a focus on actual system design and implementation. A major goal is to encourage and enable the Internet community to apply, deploy, and advance the state of available security technologies.

Our thanks to the Network and Distributed System Security (NDSS) Symposium for publishing their Creators, Authors and Presenter’s superb NDSS Symposium 2025 Conference content on the Organizations' YouTube Channel.

Permalink

The post NDSS 2025 – Diffence: Fencing Membership Privacy With Diffusion Models appeared first on Security Boulevard.

A vulnerability, which was classified as critical, was found in yued-fe LuLu UI up to 3.0.0. This issue affects the function child_process.exec of the file run.js. The manipulation results in os command injection. This vulnerability is reported as CVE-2026-2544. The attack can be launched remotely. No exploit exists. The vendor was contacted early about this disclosure but did not respond in any way.

今日暂未更新资讯~
更多最新文章，请访问SecWiki

A vulnerability, which was classified as critical, has been found in vichan-devel vichan up to 5.1.5. This vulnerability affects unknown code of the file inc/mod/pages.php of the component Password Change Handler. The manipulation of the argument Password leads to unverified password change. This vulnerability is documented as CVE-2026-2543. The attack can be initiated remotely. There is not any exploit available. The vendor was contacted early about this disclosure but did not respond in any way.

Submit #749722 / VDB-346153

A vulnerability classified as problematic was found in ArangoDB Community Edition 3.4.2-1. This affects an unknown part of the file /_db/_system/_admin/aardvark/index.html of the component Aardvark Web Admin Interface. Executing a manipulation can lead to cross site scripting. This vulnerability is registered as CVE-2019-25367. It is possible to launch the attack remotely. Furthermore, an exploit is available.

A vulnerability classified as problematic has been found in OPNsense 19.1. Affected by this issue is some unknown functionality of the component Proxy Endpoint. Performing a manipulation of the argument ignoreLogACL results in cross site scripting. This vulnerability is cataloged as CVE-2019-25376. It is possible to initiate the attack remotely. Furthermore, there is an exploit available.