<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<id>https://aisciencetracker.org/feed.xml</id><title>AI Science Tracker: additions and source checks</title><link rel="self" href="https://aisciencetracker.org/feed.xml"/><link href="https://aisciencetracker.org/"/><updated>2026-11-17T00:00:00Z</updated><author><name>AI Science Tracker</name></author>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112319</id><title>MIT krb5 kadmind XDR decoding permits authenticated memory exhaustion</title><link href="https://aisciencetracker.org/security.html#cve-2026-112319"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>In the RPCSEC_GSS unwrap path, attacker-controlled XDR length values can trigger large allocations before MIC verification; some decode-failure paths also leave allocations unreleased. A valid RPCSEC_GSS security context and access to a service using this path are required. Repeated requests can… (Vulnerability - AISLE / Red Hat Product Security; announced 2026-11-12)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ant-2026-cwjxx1bl</id><title>HarfBuzz avar2 breakpoint processing can amplify subset CPU work</title><link href="https://aisciencetracker.org/security.html#ant-2026-cwjxx1bl"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>avar2 subsetting processes breakpoints quadratically during offset compensation. The maintainer reports a 131 KB font taking 42 seconds of CPU, with work repeated on serializer retries. (Vulnerability - Anthropic / HarfBuzz maintainers; announced 2026-11-12)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ant-2026-3ho8jddv</id><title>HarfBuzz avar2 zero-width variation rows can amplify subset memory</title><link href="https://aisciencetracker.org/security.html#ant-2026-3ho8jddv"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A small font with many avar breakpoints and a 65,535-row region-less VarData caused dense tuple allocation; the maintainer reports a 20 KB input reaching 2.8 GB peak memory. (Vulnerability - Anthropic / HarfBuzz maintainers; announced 2026-11-12)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ant-2026-zcmleqfj</id><title>HarfBuzz contextual-class subset traversal can amplify CPU work</title><link href="https://aisciencetracker.org/security.html#ant-2026-zcmleqfj"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ContextFormat2/ChainContextFormat2 subset closure and dependency traversal repeatedly walked a large ClassDef for many classes sharing it, causing quadratic work. Maintainer fix computes intersecting classes once and adds a regression font. (Vulnerability - Anthropic / HarfBuzz maintainers; announced 2026-11-11)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ant-2026-njzq4xfp</id><title>HarfBuzz aliased gvar offsets can amplify avar2 subset resources</title><link href="https://aisciencetracker.org/security.html#ant-2026-njzq4xfp"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Malformed overlapping gvar offsets let many glyphs refer to one block; avar2 subsetting parsed, rewrote and retained a heap copy per glyph and retried. The maintainer reports a 251 KB input reaching 1.3 GiB and 26 seconds CPU. (Vulnerability - Anthropic / HarfBuzz maintainers; announced 2026-11-11)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ghsa-j5pq-cm28-qm6v</id><title>FreeRDP shadow server accepts unauthenticated Fast-Path input before session activation</title><link href="https://aisciencetracker.org/security.html#ghsa-j5pq-cm28-qm6v"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Before a shadow server reaches the authenticated active-session state, its Fast-Path input handlers forward mouse, keyboard and synchronization events to the X11 backend without the state guard used by the slow-path handlers. A network client can negotiate TLS-only RDP and send Fast-Path input… (Vulnerability - FreeRDP maintainers / reporter; announced 2026-11-09)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112321</id><title>Krb5: mit krb5 kadmind unauthenticated memory leak dos via auth_gssapi_init error paths</title><link href="https://aisciencetracker.org/security.html#cve-2026-112321"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>The AUTH_GSSAPI handler in lib/rpc/svc_auth_gssapi.c in MIT Kerberos 5 (aka krb5) does not free decoded arguments on certain error paths, which allows remote attackers to cause a denial of service (memory exhaustion in kadmind) via crafted AUTH_GSSAPI_INIT requests that trigger authentication… (Vulnerability - AISLE; Red Hat; announced 2026-11-09)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-74841</id><title>Apache OFBiz: Denial of Service via unauthenticated userTenantId in multitenant mode</title><link href="https://aisciencetracker.org/security.html#cve-2026-74841"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>In multitenant deployments, ContextFilter and LoginWorker built a per-tenant delegator name from a request-supplied userTenantId parameter and requested a delegator for it before confirming the tenant existed. DelegatorFactory&#x27;s internal cache retains every delegator name it is asked to resolve… (Vulnerability - AISLE (self-attributed AI-engine discovery); Apache Software Foundation CNA record; announced 2026-11-05)</summary></entry>
<entry><id>https://aisciencetracker.org/share/2026-11-ml-guided-yadh-nonenal.html</id><title>Machine learning guides yeast alcohol dehydrogenase specificity for trans-2-nonenal</title><link href="https://aisciencetracker.org/share/2026-11-ml-guided-yadh-nonenal.html"/><updated>2026-11-17T00:00:00Z</updated><category term="discovery"/><summary>Researchers measured a training library of yeast alcohol dehydrogenase I variants, used support-vector activity regression/classification and a thermostability model to rank combinations, then tested 22 prioritized variants. Four showed higher specific activity toward trans-2-nonenal than wild… (Discovery - RIKEN Center for Sustainable Resource Science; Graduate School of Science, Technology, and Innovation, Kobe University; announced 2026-11-01)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ghsa-3cp4-779m-h837</id><title>Claude-assisted discovery of a dangling channel pointer in OpenEXRCore</title><link href="https://aisciencetracker.org/security.html#ghsa-3cp4-779m-h837"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>An error path in exr_copy_unset_attributes can leave a dangling part-&gt;channels pointer after freeing channel data. Under the documented ordering condition, later error handling can obscure the original failure. The advisory limits applicability to applications using the OpenEXRCore API; it does… (Vulnerability - Academy Software Foundation / Ada Logics / Anthropic; announced 2026-10-31)</summary></entry>
<entry><id>https://aisciencetracker.org/share/2026-10-d2ml-guided-ros1-selective-compound-optimization.html</id><title>D2ML guided prospective ROS1 potency and selectivity optimization</title><link href="https://aisciencetracker.org/share/2026-10-d2ml-guided-ros1-selective-compound-optimization.html"/><updated>2026-11-17T00:00:00Z</updated><category term="discovery"/><summary>Assay-aware models were used to select 84 new compounds for synthesis and prospective testing in a ROS1 kinase program, with TRKB as an anti-target. The most selective prospective compound had a measured ROS1 IC50 of 14.7 nM and 233-fold selectivity over TRKB, compared with 20.3 nM and 79-fold for… (Discovery - Kimia Therapeutics; announced 2026-10-29)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-96644</id><title>Pillow: Heap over-read in WMF/EMF rendering via 16-bit dimension truncation (Windows builds)</title><link href="https://aisciencetracker.org/security.html#cve-2026-96644"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. Prior to 13.0.0 on Windows, src/PIL/WmfImagePlugin.py passes unchecked 32-bit WMF or EMF dimensions to the native drawwmf handler in src/display.c. The handler sizes a DIB allocation from 16-bit-truncated bcWidth and bcHeight fields but copies pixel data using… (Vulnerability - AISLE (self-attributed AI-engine discovery); GitHub CNA record; announced 2026-10-21)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-114038</id><title>Device-mapper-multipath: device-mapper-multipath: information disclosure via malformed format string</title><link href="https://aisciencetracker.org/security.html#cve-2026-114038"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A flaw was found in device-mapper-multipath. A local user can trigger an out-of-bounds read by sending a command with a format string ending in a trailing percent character to the multipath daemon. (Vulnerability - AISLE; Red Hat; announced 2026-10-21)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112895</id><title>Pillow: Unbounded recursion parsing a PDF page tree or trailer chain (incomplete fix for CVE-2026-42310)</title><link href="https://aisciencetracker.org/security.html#cve-2026-112895"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. From 5.1.0 until 13.0.0, Pillow can exhaust resources while parsing an existing PDF during an append operation. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112893</id><title>Pillow: Invalid free in FreeType glyph rendering when a stroker is active</title><link href="https://aisciencetracker.org/security.html#cve-2026-112893"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. From 10.0.0 until 13.0.0, Pillow can free an invalid glyph pointer during FreeType rendering with an active stroker. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112892</id><title>Pillow: MSP RLE decoder expands a tiny file into unbounded memory</title><link href="https://aisciencetracker.org/security.html#cve-2026-112892"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. From 4.1.0 until 13.0.0, Pillow&#x27;s MSP RLE decoder can expand a small file into an unbounded in-memory buffer. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112405</id><title>Pillow: NULL pointer dereference in Image.merge via an unvalidated band 0</title><link href="https://aisciencetracker.org/security.html#cve-2026-112405"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. From 12.0.0 until 13.0.0, Pillow&#x27;s Image.merge() can dereference a null image8 plane. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112397</id><title>Pillow: Out-of-bounds access in _imagingmorph on a zero-width image</title><link href="https://aisciencetracker.org/security.html#cve-2026-112397"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. Prior to 13.0.0, Pillow&#x27;s ImageMorph.MorphOp.apply() can access memory before a row buffer for a zero-width image. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-112396</id><title>Pillow: Heap use-after-free when encoding TIFF to an in-memory buffer</title><link href="https://aisciencetracker.org/security.html#cve-2026-112396"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Pillow is a Python imaging library. From 11.3.0 until 13.0.0, Pillow has a heap use-after-free when libtiff rejects a tag while encoding TIFF data to an in-memory buffer. (Vulnerability - AISLE; announced 2026-10-19)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-85977</id><title>Libtirpc: libtirpc: denial of service via unbounded memory allocation during portmapper response decoding</title><link href="https://aisciencetracker.org/security.html#cve-2026-85977"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A flaw was found in libtirpc. A remote attacker can exploit this vulnerability by sending specially crafted PMAPPROC_DUMP replies to a client application that uses the pmap_getmaps() function against an attacker-controlled host. This unbounded decoding can lead to excessive memory and CPU… (Vulnerability - AISLE (self-attributed AI-engine discovery); Red Hat Product Security CNA record; announced 2026-10-16)</summary></entry>
<entry><id>https://aisciencetracker.org/share/2026-10-designed-metalloproteases.html</id><title>Deep-learning design of zinc metalloproteases</title><link href="https://aisciencetracker.org/share/2026-10-designed-metalloproteases.html"/><updated>2026-11-17T00:00:00Z</updated><category term="discovery"/><summary>RFdiffusion2 for Molecular Interfaces designs zinc-dependent proteases for peptide-bond hydrolysis. The primary paper reports cleavage at the targeted bond in 36% of 135 designs and selected designs that cleave disease-associated proteins in vitro. These biochemical results do not establish… (Discovery - Institute for Protein Design, University of Washington; announced 2026-10-14)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-109833</id><title>Gnome-remote-desktop: heap buffer overflow in gnome-remote-desktop vnc tls peek buffer handling allows remote denial of service</title><link href="https://aisciencetracker.org/security.html#cve-2026-109833"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A flaw was found in gnome-remote-desktop as shipped in Red Hat Enterprise Linux. In the VNC TLS read path, partial consumption of the internal peek buffer incorrectly moves remaining peeked bytes into the caller-supplied buffer instead of compacting the peek buffer. (Vulnerability - AISLE; announced 2026-10-13)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-109832</id><title>Gnome-remote-desktop: busy-wait dos in vnc tls i/o handlers (gnutls_e_again / gnutls_e_interrupted)</title><link href="https://aisciencetracker.org/security.html#cve-2026-109832"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A flaw was found in gnome-remote-desktop as shipped in Red Hat Enterprise Linux. In the VNC TLS I/O handlers, retryable GnuTLS results (GNUTLS_E_AGAIN and GNUTLS_E_INTERRUPTED) are handled with an unbounded busy loop instead of yielding to the event loop. (Vulnerability - AISLE; announced 2026-10-13)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#ghsa-rh6f-3x46-j33q</id><title>vLLM shared cache keys allow cross-request state aliasing</title><link href="https://aisciencetracker.org/security.html#ghsa-rh6f-3x46-j33q"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Caller-controlled EXIF image identifiers and untyped prefix-cache extra keys can produce collisions between requests. The project advisory groups these two cache-identity sites together; the reported result is cross-request cached-state confusion, not demonstrated code execution. (Vulnerability - Patch the Planet (OpenAI and Trail of Bits); announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-82531</id><title>Smarty before 4.5.8 and 5.x before 5.8.5 PHP Code Injection via extends: Inheritance Cache</title><link href="https://aisciencetracker.org/security.html#cve-2026-82531"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Smarty before 4.5.8 and 5.x before 5.8.5 contains a code injection vulnerability where the top-level nocache_hash is never restored during extends:/multi-component template inheritance, leaving it null. Attackers can supply assigned data containing a forged SmartyNocache marker that is copied… (Vulnerability - AISLE (self-attributed AI-engine discovery); VulnCheck CNA record; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106116</id><title>ImageSharp: BigTIFF IFD count can keep a decoder thread in a non-progressing loop</title><link href="https://aisciencetracker.org/security.html#cve-2026-106116"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 2.0.0 until 4.1.2, ExifReader.ReadValues64 trusts the 64-bit BigTIFF IFD entry count and iterates once per declared entry. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106115</id><title>ImageSharp: TIFF CCITT T6 encoder writes beyond an undersized output buffer</title><link href="https://aisciencetracker.org/security.html#cve-2026-106115"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 2.1.0 until 4.1.2, the TIFF CCITT Group 4 encoder allocates Width times rowsPerStrip bytes even though T6BitCompressor.CompressStrip can emit encoded row data and two 12-bit end-of-facsimile-block codes beyond that capacity. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106114</id><title>ImageSharp: ICC CLUT parsing allocates from unvalidated channel and grid dimensions</title><link href="https://aisciencetracker.org/security.html#cve-2026-106114"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 1.0.0-beta0001 until 4.1.2, ICC CLUT parsing calculates allocation sizes from attacker-declared channel and grid dimensions before confirming that the profile contains the declared values. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106113</id><title>ImageSharp: HistogramEqualization uses an unvalidated luminance as an unchecked histogram index</title><link href="https://aisciencetracker.org/security.html#cve-2026-106113"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 2.0.0 until 4.1.2, decoding an attacker-supplied 32-bit floating-point TIFF as Image&lt;HalfVector4&gt; and applying HistogramEqualization can produce a non-finite or out-of-range luminance in ColorNumerics.GetBT709Luminance. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106112</id><title>ImageSharp: ICC LUT16 output channel count can write beyond Vector4</title><link href="https://aisciencetracker.org/security.html#cve-2026-106112"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 4.0.0 until 4.1.2, ICC LUT16 conversion accepts more than four output channels even though ClutCalculator.Calculate and LutEntryCalculator.CalculateLut store intermediate and output values in Vector4. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106111</id><title>ImageSharp: EXR ZIP decoder can expose stale allocator data after a short inflate</title><link href="https://aisciencetracker.org/security.html#cve-2026-106111"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 4.0.0 until 4.1.2, ExrBaseDecompressor.UndoZipCompression accepts a nonempty ZIP or ZIPS inflate result that is shorter than the EXR block&#x27;s required size. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-106110</id><title>ImageSharp: TIFF CCITT T4 encoder can write past its compressed output buffer</title><link href="https://aisciencetracker.org/security.html#cve-2026-106110"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>ImageSharp is a 2D graphics library. From 2.0.0 until 4.1.2, the TIFF CCITT Group 3 encoder allocates an undersized compressed-data buffer for narrow 1-bit images. (Vulnerability - AISLE; announced 2026-10-06)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-104030</id><title>Sssd: sssd: denial of service via out-of-bounds read during passkey parsing</title><link href="https://aisciencetracker.org/security.html#cve-2026-104030"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>A flaw was found in sssd. This vulnerability allows a local user to cause a Denial of Service (DoS) by submitting a specially crafted passkey authentication token that lacks null terminators. (Vulnerability - AISLE; Red Hat; announced 2026-10-05)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-71885</id><title>MLS X.509 credential not bound to the LeafNode signature key</title><link href="https://aisciencetracker.org/security.html#cve-2026-71885"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>BC-JAVA’s MLS implementation checked a LeafNode signature against its own signature key but did not verify that an attached X.509 certificate used the same key. In deployments accepting external commits without a separate credential-admission check, an attacker could use another party’s… (Vulnerability - AISLE (self-attributed AI-engine discovery); Bouncy Castle CNA record; announced 2026-10-03)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-17507</id><title>Scrutineer and Claude-assisted discovery of Bouncy Castle MLS leaf-index denial of service</title><link href="https://aisciencetracker.org/security.html#cve-2026-17507"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>The MLS implementation stores RFC 9420’s uint32 leaf_index in a signed int and compares it to the group leaf count as signed. A high-bit wire value becomes negative and passes the bounds check. In the hasLeaf path, any current group member can send an unprotected PrivateMessage whose out-of-range… (Vulnerability - Alpha-Omega / Legion of the Bouncy Castle; announced 2026-10-02)</summary></entry>
<entry><id>https://aisciencetracker.org/share/2026-10-metasynbgc-ai-prioritized-metagenomic-compounds.html</id><title>Deep-learning prioritization of metagenomic products yielded cytotoxic compounds</title><link href="https://aisciencetracker.org/share/2026-10-metasynbgc-ai-prioritized-metagenomic-compounds.html"/><updated>2026-11-17T00:00:00Z</updated><category term="discovery"/><summary>MetasynBGC used gene-cluster-family-guided metagenomic mining and a Chemprop graph neural network to prioritize predicted natural-product structures for synthesis. Six selected NRPS-derived compounds showed in-vitro cytotoxic activity across seven cancer cell lines, with reported IC50 values as… (Discovery - Shandong University; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences; Jining Medical University; announced 2026-10-01)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-93546</id><title>Apache HTTP Server: mod_dav_fs namespace overflow</title><link href="https://aisciencetracker.org/security.html#cve-2026-93546"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>Integer overflow in mod_dav_fs in Apache HTTP Server through 2.4.68 allows an authenticated WebDAV client with write access to crash worker processes and persistently corrupt a directory&#x27;s property database via PROPPATCH requests declaring many XML namespaces. (Vulnerability - AISLE (self-attributed AI-engine discovery); Apache Software Foundation CNA record; announced 2026-10-01)</summary></entry>
<entry><id>https://aisciencetracker.org/share/2026-09-pca-map-antiplasmodial-haemozoin-screen.html</id><title>PCA mapping and machine learning prioritize antiplasmodial chemotypes</title><link href="https://aisciencetracker.org/share/2026-09-pca-map-antiplasmodial-haemozoin-screen.html"/><updated>2026-11-17T00:00:00Z</updated><category term="discovery"/><summary>A PCA enrichment map and machine-learning virtual screening pipeline prioritized a commercial library for beta-haematin and parasite assays. The primary indexed paper reports several chemotypes evaluated for antiplasmodial activity, beta-haematin inhibition and haemozoin formation inhibition. This… (Discovery - University of Cape Town; announced 2026-09-29)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-102497</id><title>Apache XMLSchema walker recurses on cyclic schema definitions</title><link href="https://aisciencetracker.org/security.html#cve-2026-102497"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>The xmlschema-walker fails to detect cycles in type derivation, substitution groups, model groups, or attribute groups. A malicious schema containing such a cycle can recurse until the stack overflows and cause denial of service. The public sources do not specify a network protocol or establish a… (Vulnerability - Anthropic / Apache XMLSchema maintainers; announced 2026-09-29)</summary></entry>
<entry><id>https://aisciencetracker.org/security.html#cve-2026-102496</id><title>Apache XMLSchema nested schema structures exhaust the model builder stack</title><link href="https://aisciencetracker.org/security.html#cve-2026-102496"/><updated>2026-11-17T00:00:00Z</updated><category term="vulnerability"/><summary>The XMLSchema model builder does not limit deeply nested schema structures. A malicious schema with excessive nesting can recurse until the stack overflows and cause denial of service. The public sources do not specify a network protocol or establish a particular downstream product deployment. (Vulnerability - Anthropic / Apache XMLSchema maintainers; announced 2026-09-29)</summary></entry>
</feed>
