# Reverse Engineering

Reverse engineering is the process of analyzing existing software, hardware, or other systems to recover their design, structure, and function, often using disassembly, decompilation, and dynamic analysis.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## AI Slop Reduction Policy

DevFeed: [AI Slop Reduction Policy](<https://devfeed.tech/articles/ai-slop-reduction-policy-146650.md>)

Original publisher: [Read original article](<https://libreboot.org/news/clankers.html>)

Author: Lhehaa Rhoeow

Published: 2026-11-20T20:58:00.351381Z

Content type: opinion

Language: en

Sources: [News about Libreboot releases and development](<https://devfeed.tech/sources/news-about-libreboot-releases-and-development.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [ai slop pull requests](<https://devfeed.tech/topics/ai-slop-pull-requests.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [libreboot](<https://devfeed.tech/topics/libreboot.md>), [coreboot](<https://devfeed.tech/topics/coreboot.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [ai-code](<https://devfeed.tech/tags/ai-code.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [ai-policy](<https://devfeed.tech/tags/ai-policy.md>), [ai-slop](<https://devfeed.tech/tags/ai-slop.md>), [ai-use](<https://devfeed.tech/tags/ai-use.md>), [article](<https://devfeed.tech/tags/article.md>), [bios](<https://devfeed.tech/tags/bios.md>), [coreboot](<https://devfeed.tech/tags/coreboot.md>), [free-software](<https://devfeed.tech/tags/free-software.md>), [libre](<https://devfeed.tech/tags/libre.md>), [libreboot](<https://devfeed.tech/tags/libreboot.md>), [news](<https://devfeed.tech/tags/news.md>), [opensource](<https://devfeed.tech/tags/opensource.md>), [policy](<https://devfeed.tech/tags/policy.md>), [reduction](<https://devfeed.tech/tags/reduction.md>), [slop](<https://devfeed.tech/tags/slop.md>), [uefi](<https://devfeed.tech/tags/uefi.md>)

### AI overview

Libreboot's policy is to reduce AI-generated code where feasible, while acknowledging it cannot remove all such code from upstream projects or promise a total ban. The author says Libreboot distributes AI-generated code through coreboot, and describes reverting an optional battery-threshold patch and removing thinkpad_acpi support after a fix broke WWAN/Bluetooth. The policy generally recommends avoiding AI and expects most AI-generated patches to be rejected, but leaves room to accept work judged sound when it advances free-software support or replaces proprietary blobs. It treats AI-assisted reverse engineering as a possible short-term aid requiring skilled human understanding, while warning about deskilling, review burdens, and scraping traffic that can disrupt small projects. The author plans to merge a trusted engineer's AI-assisted Intel FSP replacement and continues to recommend ReBIOS, illustrating that code quality and replacing proprietary software can outweigh its development method in specific cases.

### Source excerpt

Article: AI Slop Reduction Policy Web link: https://libreboot.org/news/clankers.html

## LLM assisted reverse engineering

DevFeed: [LLM assisted reverse engineering](<https://devfeed.tech/articles/llm-assisted-reverse-engineering-144490.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/19/llm-assisted-reverse-engineering/>)

Author: Anne Barela

Published: 2026-11-19T22:17:09Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Ghidra](<https://devfeed.tech/topics/ghidra.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-coding-agent](<https://devfeed.tech/tags/ai-coding-agent.md>), [assembly](<https://devfeed.tech/tags/assembly.md>), [csv](<https://devfeed.tech/tags/csv.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [ghidra](<https://devfeed.tech/tags/ghidra.md>), [github](<https://devfeed.tech/tags/github.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [log](<https://devfeed.tech/tags/log.md>), [markdown-files](<https://devfeed.tech/tags/markdown-files.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [programming](<https://devfeed.tech/tags/programming.md>), [radare2](<https://devfeed.tech/tags/radare2.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [skills](<https://devfeed.tech/tags/skills.md>), [software](<https://devfeed.tech/tags/software.md>)

### AI overview

The author uses an AI coding agent to analyze binaries and produce code overviews that would take much longer to assemble by hand from assembly. One published skill documents tools such as radare2 and a Ghidra MCP bridge, with 22 scripts for targeted decompilation; it addresses agents stopping early because they do not know how to run disassembly. A second skill makes a stuck agent state one testable hypothesis at a time and log results in a CSV loop, helping prevent random guessing and circular work. Findings are recorded in Markdown as the analysis proceeds.

### Source excerpt

ioma8 has been taking apart hardware and software for two decades, and in a post on slowchronicle.com describes how that work now runs with an AI coding agent doing the reading. The agent is pointed at the binary, along with a set of purpose-written skills, and it produces an overview of the code that would [...]

## Digital Badge Gets Reverse Engineered

DevFeed: [Digital Badge Gets Reverse Engineered](<https://devfeed.tech/articles/digital-badge-gets-reverse-engineered-144156.md>)

Original publisher: [Read original article](<https://hackaday.com/2026/11/19/digital-badge-gets-reverse-engineered/>)

Author: Zoe Skyforest

Published: 2026-11-19T18:00:00Z

Content type: news

Language: en

Sources: [Hackaday](<https://devfeed.tech/sources/hackaday.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>), [Ghidra](<https://devfeed.tech/topics/ghidra.md>)

Tags: [apk](<https://devfeed.tech/tags/apk.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [badge](<https://devfeed.tech/tags/badge.md>), [bluetooth](<https://devfeed.tech/tags/bluetooth.md>), [bluetooth-low-energy](<https://devfeed.tech/tags/bluetooth-low-energy.md>), [buttons](<https://devfeed.tech/tags/buttons.md>), [capacitive](<https://devfeed.tech/tags/capacitive.md>), [capacitive-touch](<https://devfeed.tech/tags/capacitive-touch.md>), [controller](<https://devfeed.tech/tags/controller.md>), [custom](<https://devfeed.tech/tags/custom.md>), [custom-firmware](<https://devfeed.tech/tags/custom-firmware.md>), [develop](<https://devfeed.tech/tags/develop.md>), [device](<https://devfeed.tech/tags/device.md>), [digital](<https://devfeed.tech/tags/digital.md>), [display](<https://devfeed.tech/tags/display.md>), [download](<https://devfeed.tech/tags/download.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [flash](<https://devfeed.tech/tags/flash.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [misc-hacks](<https://devfeed.tech/tags/misc-hacks.md>), [product](<https://devfeed.tech/tags/product.md>), [reverse](<https://devfeed.tech/tags/reverse.md>)

### AI overview

The $10 JieLi AC707N digital badge can run custom firmware, enabling uses such as a Bluetooth mouse or scanner. Reverse engineering began with a physical teardown and decompilation of the companion APK in Ghidra; this enabled downloading firmware without authentication and developing an open SDK. The badge has a 360x360 QSPI display, AXS5106L capacitive touch controller, two buttons, 4 MB of flash, and Bluetooth Low Energy. The excerpt says custom firmware is available on GitHub and can be extended for other uses.

### Source excerpt

The JieLi AC707N is a cheap $10 digital badge that actually packs in a fair bit of interesting hardware. [Sabas] reverse-engineered the product and figured out how to flash custom ...read more

## Lodestone: Automated Semantic Discovery of Datacenter Hardware Specialization

DevFeed: [Lodestone: Automated Semantic Discovery of Datacenter Hardware Specialization](<https://devfeed.tech/articles/lodestone-automated-semantic-discovery-of-datacenter-hardware-specialization-142687.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23663>)

Author: Peizhe Liu, Akanksha Jain, Gerasimos Gerogiannis, Vidushi Dadu, Parthasarathy Ranganathan, Jovan Stojkovic

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [RISC-V](<https://devfeed.tech/topics/riscv.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>)

Tags: [accelerators](<https://devfeed.tech/tags/accelerators.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [applications](<https://devfeed.tech/tags/applications.md>), [automated](<https://devfeed.tech/tags/automated.md>), [code](<https://devfeed.tech/tags/code.md>), [control](<https://devfeed.tech/tags/control.md>), [core](<https://devfeed.tech/tags/core.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [datacenter](<https://devfeed.tech/tags/datacenter.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experts](<https://devfeed.tech/tags/experts.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [specialization](<https://devfeed.tech/tags/specialization.md>)

### AI overview

Lodestone uses semantic reasoning to find hardware specialization opportunities that hot-function or instruction-pattern analysis may miss when similar computations are spread across applications and libraries. It extracts hot control-flow regions from execution traces, packages them for agent analysis, groups regions that compute the same thing, and checks semantic correctness iteratively. It then tests each family against reference inputs and outputs from recorded executions. For selected families, it proposes hardware such as ISA extensions, near-core units, or detached accelerators; human experts choose which proposals to pursue. Across 20 workloads in six datacenter categories, 30 discovered families were simulated and implemented in RTL. The reported optimizations sped workloads up by 12.6% on average, with a maximum of 80.2%.

### Source excerpt

arXiv:2611.23663v1 Announce Type: cross Abstract: Hardware specialization can improve datacenter efficiency. Unfortunately, identifying computations that justify dedicated hardware is difficult. Existing approaches analyze hot functions, instruction patterns, or overheads such as datacenter tax. They can miss fine-grained computations whose aggregate cost is spread across applications and libraries, especially when different pieces of code obscure the same underlying computation. We observe that seemingly unrelated code regions often perform the same computations, creating specialization opportunities that require semantic reasoning. The rise of agentic AI makes it possible to automate this. However, conventional AI agents cannot efficiently and reliably analyze large scale datacenter software. To unlock this opportunity, we present Lodestone, an agentic framework that automatically discovers hardware specialization opportunities by reasoning about what the executed code computes. First, Lodestone extracts hot control-flow regions from execution traces across workloads and organizes them into compact analysis packages that make large workloads tractable for agents. Second, agents describe what each region computes, merge those that perform the same computation into computational families, and check their semantic correctness in a loop. Third, to verify the discovered family, Lodestone runs every piece of code in a family against a set of reference inputs and outputs on recorded executions. Finally, for selected families, Lodestone proposes hardware, e.g., ISA extensions, near-core units, or detached accelerators, and human experts decide which proposals to develop further. We evaluate Lodestone on 20 workloads across six datacenter workload categories. We simulate 30 discovered families in a cycle-accurate simulator and implement them in RTL. These optimizations speed up the workloads by 12.6% on average and by up to 80.2%.

## The latest edition of Phrack magazine is out

DevFeed: [The latest edition of Phrack magazine is out](<https://devfeed.tech/articles/the-latest-edition-of-phrack-magazine-is-out-141068.md>)

Original publisher: [Read original article](<https://blog.adafruit.com/2026/11/18/the-latest-edition-of-phrack-magazine-is-out/>)

Author: Anne Barela

Published: 2026-11-18T19:22:01Z

Content type: news

Language: en

Sources: [Adafruit Industries](<https://devfeed.tech/sources/adafruit-industries-makers-hackers-artists-designers-and-engineers.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Memory Safety](<https://devfeed.tech/topics/memory-safety.md>)

Tags: [backdoors](<https://devfeed.tech/tags/backdoors.md>), [binary-ninja](<https://devfeed.tech/tags/binary-ninja.md>), [botnets](<https://devfeed.tech/tags/botnets.md>), [boxing](<https://devfeed.tech/tags/boxing.md>), [chromium](<https://devfeed.tech/tags/chromium.md>), [computers](<https://devfeed.tech/tags/computers.md>), [container](<https://devfeed.tech/tags/container.md>), [exploitation](<https://devfeed.tech/tags/exploitation.md>), [halvar-flake](<https://devfeed.tech/tags/halvar-flake.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [history](<https://devfeed.tech/tags/history.md>), [interview](<https://devfeed.tech/tags/interview.md>), [issue](<https://devfeed.tech/tags/issue.md>), [kernel](<https://devfeed.tech/tags/kernel.md>), [linux-kernel](<https://devfeed.tech/tags/linux-kernel.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [machine-learning-models](<https://devfeed.tech/tags/machine-learning-models.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [phrack](<https://devfeed.tech/tags/phrack.md>), [publications](<https://devfeed.tech/tags/publications.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>)

### AI overview

Phrack's third issue under its new staff contains 23 main papers and eight shorter Line Noise articles, making it larger than last year's 40th anniversary issue. Topics include exploitation, reverse engineering, machine learning, hardware, and hacking history. Technical papers cover vDSO container escapes, use-after-free exploitation, QNX shellcode, Linux kernel research, a Chromium Skia sandbox escape, Windows smart card architecture, Binary Ninja static taint analysis, underwater vehicles, and machine learning backdoors. Other pieces trace WinFSP zero-day bugs to PhrackCTF challenges and recount episodes from hacking history, including the SSH1 CRC32 vulnerability featured in The Matrix Reloaded. The issue also includes an interview with Halvar Flake. Editors invite readers to submit write-ups for the next issue.

### Source excerpt

Phrack has published a new issue, the third put together by the magazine's new staff. It's larger than last year's 40th anniversary release, with 23 main papers and eight shorter articles in Line Noise. The subjects run from exploitation and reverse engineering to machine learning, hardware, and the history and culture around this work. The [...]

## Building Software Got Faster. Building Trust Didn't.

DevFeed: [Building Software Got Faster. Building Trust Didn't.](<https://devfeed.tech/articles/building-software-got-faster-building-trust-didn-t-141948.md>)

Original publisher: [Read original article](<https://spin.atomicobject.com/building-software-ai-risk/>)

Author: Jason Porritt

Published: 2026-11-18T13:00:38Z

Content type: article

Language: en

Sources: [Atomic Object](<https://devfeed.tech/sources/atomic-object.md>)

Topics: [trust](<https://devfeed.tech/topics/trust.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Extension](<https://devfeed.tech/topics/extension.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-tool](<https://devfeed.tech/tags/ai-tool.md>), [app](<https://devfeed.tech/tags/app.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [browser](<https://devfeed.tech/tags/browser.md>), [browser-extension](<https://devfeed.tech/tags/browser-extension.md>), [building](<https://devfeed.tech/tags/building.md>), [code](<https://devfeed.tech/tags/code.md>), [custom](<https://devfeed.tech/tags/custom.md>), [custom-firmware](<https://devfeed.tech/tags/custom-firmware.md>), [desk](<https://devfeed.tech/tags/desk.md>), [device](<https://devfeed.tech/tags/device.md>), [extension](<https://devfeed.tech/tags/extension.md>), [faster](<https://devfeed.tech/tags/faster.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [first](<https://devfeed.tech/tags/first.md>), [it](<https://devfeed.tech/tags/it.md>), [know](<https://devfeed.tech/tags/know.md>), [risk](<https://devfeed.tech/tags/risk.md>), [software](<https://devfeed.tech/tags/software.md>), [the-software-life](<https://devfeed.tech/tags/the-software-life.md>), [trust](<https://devfeed.tech/tags/trust.md>)

### AI overview

AI tools have made bespoke software and firmware much faster to build, while review and trust practices have not kept pace. The excerpt cites examples ranging from one-evening personal tools to more than 130 custom firmwares for a synthesizer within five weeks, alongside reports of serious flaws and unreviewed code. The risk reaches organizations through staff-built or downloaded tools, customers using custom interfaces or agents, and suppliers' products and components. The article describes threats including malicious add-ons, compromised open-source code, deceptive browser extensions, and AI agents following hidden web-page instructions. It recommends maintaining traditional security practices while giving staff a timely path to propose experiments, deciding how openly to support customer-built integrations, and defining how to handle unusual access patterns. These are decisions to make; the excerpt does not prescribe specific approval or monitoring controls.

### Source excerpt

Somewhere in your organization, someone has built or downloaded a piece of software you do not know about. It might be a browser extension, a small app that automates a report, or custom firmware on a device sitting on their desk. It was probably built with an AI tool, and nobody has checked the code, [...] The post Building Software Got Faster. Building Trust Didn't. appeared first on Atomic Spin.

## Dissecting the GPU Load/Store Unit and Exposing Its Security Risks

DevFeed: [Dissecting the GPU Load/Store Unit and Exposing Its Security Risks](<https://devfeed.tech/articles/dissecting-the-gpu-load-store-unit-and-exposing-its-security-risks-141504.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23022>)

Author: Wenbo Han, Pierre Talbot

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Security](<https://devfeed.tech/topics/security.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Hopper](<https://devfeed.tech/topics/hopper.md>)

Tags: [aes](<https://devfeed.tech/tags/aes.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attack-surface](<https://devfeed.tech/tags/attack-surface.md>), [bank](<https://devfeed.tech/tags/bank.md>), [channel](<https://devfeed.tech/tags/channel.md>), [costs](<https://devfeed.tech/tags/costs.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cublas](<https://devfeed.tech/tags/cublas.md>), [defense](<https://devfeed.tech/tags/defense.md>), [global](<https://devfeed.tech/tags/global.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [h100](<https://devfeed.tech/tags/h100.md>), [latency](<https://devfeed.tech/tags/latency.md>), [leak](<https://devfeed.tech/tags/leak.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [matrix](<https://devfeed.tech/tags/matrix.md>), [memory](<https://devfeed.tech/tags/memory.md>), [mps](<https://devfeed.tech/tags/mps.md>), [risks](<https://devfeed.tech/tags/risks.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

Microbenchmarks of recent NVIDIA GPUs yield a cycle-accurate latency formula for global L1 hits and shared-memory loads, with address-dependent costs from bank conflicts and L1 tag lookups. The study finds contention at both points and shows that a full LSU pipeline can let one warp starve a warp from a later-launched block. These mechanisms expand the reported attack surface: either bank conflicts or tag lookups alone can recover an AES-128 key under the threat model of prior GPU AES timing attacks. Under MPS, each also enables a cross-process covert channel whose rate rises across GPU generations. LSU starvation can slow colocated cuBLAS matrix multiplication by up to 750x on A100 and 563x on H100. The supplied excerpt reports these risks and measurements but gives no defenses.

### Source excerpt

arXiv:2611.23022v1 Announce Type: cross Abstract: GPUs are the main platform for large-scale machine learning and power the largest supercomputers. Their load/store units (LSUs) often determine kernel performance and have been exploited in prior attacks. Yet what is known about the LSU is only a few isolated mechanisms and parameters, and even a single one has been enough for an attack. In particular, it is still unknown what a load costs, where loads contend, and even whether a warp can starve due to contention. Until each is measured, neither its security risk nor a defense is known. We reverse-engineer how the LSU serves load instructions on recent NVIDIA GPUs with microbenchmarks and analyze the security implications. We give a cycle-accurate latency formula for a global L1 hit or shared memory access from the addresses it touches, whose address-dependent terms are bank conflicts and the L1 tag lookup; we show that loads contend at both, and that when the LSU pipeline is full a warp can starve a warp of a later-launched block. We then show that the LSU's attack surface is far broader than prior attacks revealed: through channels never exploited before, it can leak information, communicate covertly, and degrade colocated workloads. Through either global bank conflicts or tag lookups alone, we recover a full AES-128 key under the threat model of prior GPU AES timing attacks. Under NVIDIA's Multi-Process Service (MPS), the bank access and the tag lookup each open a covert channel across processes, whose rate rises with each GPU generation. Also under MPS, this starvation lets one warp slow a colocated cuBLAS matrix multiplication by up to 750x on the A100 and 563x on the H100.

## Lasso: Towards Sustainable AI-Generated Media Detection with Intel QAT On-Chip Compression

DevFeed: [Lasso: Towards Sustainable AI-Generated Media Detection with Intel QAT On-Chip Compression](<https://devfeed.tech/articles/lasso-towards-sustainable-ai-generated-media-detection-with-intel-qat-on-chip-compression-137360.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18895>)

Author: Darsh Asher, Joshua Kalyanapu, Kaushal Mhapsekar, Bita Aslrousta, Amna Hassan, Debayan Sengupta, Samira Mirbagher Ajorpaz

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Sustainable Computing](<https://devfeed.tech/topics/sustainable-computing.md>), [Compression](<https://devfeed.tech/topics/compression.md>), [Generative AI](<https://devfeed.tech/topics/generative-ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-generated-content](<https://devfeed.tech/tags/ai-generated-content.md>), [ai-generated-media](<https://devfeed.tech/tags/ai-generated-media.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [binoculars](<https://devfeed.tech/tags/binoculars.md>), [chip](<https://devfeed.tech/tags/chip.md>), [code](<https://devfeed.tech/tags/code.md>), [compression](<https://devfeed.tech/tags/compression.md>), [content](<https://devfeed.tech/tags/content.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ar](<https://devfeed.tech/tags/cs-ar.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [data](<https://devfeed.tech/tags/data.md>), [data-structure](<https://devfeed.tech/tags/data-structure.md>), [detection](<https://devfeed.tech/tags/detection.md>), [device](<https://devfeed.tech/tags/device.md>), [domains](<https://devfeed.tech/tags/domains.md>), [edge](<https://devfeed.tech/tags/edge.md>), [energy](<https://devfeed.tech/tags/energy.md>), [intel](<https://devfeed.tech/tags/intel.md>), [media](<https://devfeed.tech/tags/media.md>)

### AI overview

Lasso uses timing differences in Intel QuickAssist Technology's compression accelerator as a signal for detecting AI-generated media. For text, the abstract reports 100% accuracy on original DetectRL text, 92% average after paraphrasing, and 82% across original, paraphrased, and perturbed text; accuracy reaches 95% for matched instruction-tuned generators, including an unseen generator. On RAID, it averages 82% when tested on a different generator and 74% across eight held-out domains. Detection takes 0.74 microseconds, at least 83,784 times faster than five baselines, and uses over 99% less energy per document than Binoculars and DetectGPT. It also reports 67.5% accuracy for video, 70% for code, and 78% for images. These are abstract-level reported results; evaluation details are not included in the supplied excerpt.

### Source excerpt

arXiv:2611.18895v1 Announce Type: cross Abstract: Synthetic text, images, and video are rising. Software-only detectors are power-hungry, too slow, and costly, and do not scale with demand. When software cannot scale, historically hardware often comes to the rescue. We perform the first reverse engineering of a compression accelerator--Intel QuickAssist Technology (QAT) microarchitecture and--find different amount of time on each, a novel data structure-dependent property. Using this information, we then build Lasso, a low-cost, high-performance detector for AI-generated content that turns the accelerator into a measuring instrument. For text detection, Lasso achieves 100% accuracy on original DetectRL text and retains 92% average accuracy after paraphrasing, compared with 78% for Binoculars and 55% for DetectGPT. Across original, paraphrased, and perturbed text, it achieves 82% average accuracy, outperforming all five baselines. With topic, length, and formatting matched, accuracy reaches 95% for instruction-tuned generators, including when the tested generator is unseen. On RAID, accuracy averages 82% when trained on one generator and tested on another, and 74% across eight held-out domains. Detection takes 0.74 microseconds, a speedup of at least 83,784x across five baselines. At the same time, Lasso consumes over 99% less energy per document than both Binoculars and DetectGPT. At the same time, Lasso detects AI-generated videos, software code, and images with 67.5%, 70% and 78% accuracy, respectively, with low latency and power opening avenues for practical real time and edge device detection of multimodality AI artifacts for the first time. No other compared technique supports detection across all four modalities: text, video, images, and code.

## Mining Architectural Models in the Wild: A Curated Dataset of AADL Models

DevFeed: [Mining Architectural Models in the Wild: A Curated Dataset of AADL Models](<https://devfeed.tech/articles/mining-architectural-models-in-the-wild-a-curated-dataset-of-aadl-models-137952.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18566>)

Author: Thi Dinh Tran, Maria Teresa Rossi, Ludovico Iovino, Leonardo Mariani

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>)

Tags: [aggregation](<https://devfeed.tech/tags/aggregation.md>), [ai](<https://devfeed.tech/tags/ai.md>), [alignment](<https://devfeed.tech/tags/alignment.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [annotation](<https://devfeed.tech/tags/annotation.md>), [annotations](<https://devfeed.tech/tags/annotations.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [component](<https://devfeed.tech/tags/component.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [design](<https://devfeed.tech/tags/design.md>), [domain](<https://devfeed.tech/tags/domain.md>), [domains](<https://devfeed.tech/tags/domains.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [formats](<https://devfeed.tech/tags/formats.md>), [mining](<https://devfeed.tech/tags/mining.md>), [models](<https://devfeed.tech/tags/models.md>)

### AI overview

A3D provides a curated dataset of 1,115 AADL models mined from public GitHub repositories and validated with OSATE, addressing the stated shortage of large, annotated architecture-model benchmarks. It covers 7 generic domains and 52 specific-problem clusters, with domain- and problem-level labels plus component mappings to three reference architectures validated through a user study. The models are available in AADL and JSON formats, alongside provenance, a reproducible mining pipeline, and a validation protocol.

### Source excerpt

arXiv:2611.18566v1 Announce Type: new Abstract: Empirical research on software architecture maintenance and evolution is hindered by the scarcity of large-scale, validated, and richly annotated model datasets. Existing benchmarks are limited in scale, provide only model-level annotations, or lack alignment with reference architectures, thereby restricting reproducible studies of architectural reuse, conformance, and AI-driven analysis. To address this gap, we present A3D, a curated dataset of 1,115 Architecture Analysis and Design Language (AADL) models mined from public GitHub repositories and validated with the OSATE tool. The dataset spans 7 generic domains and 52 specific-problem clusters, and provides two annotation layers: (i) domain- and problem-level labels, which capture different model aggregation criteria, and (ii) component-level mappings to three reference architectures, which have been externally validated through a user study. Models are released in AADL and JSON formats with full provenance, a reproducible mining pipeline, and the validation protocol.

## Revisiting Re-executability: Coverage-Guided Symbolic Feedback to Improve Decompilation

DevFeed: [Revisiting Re-executability: Coverage-Guided Symbolic Feedback to Improve Decompilation](<https://devfeed.tech/articles/revisiting-re-executability-coverage-guided-symbolic-feedback-to-improve-decompilation-137973.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18892>)

Author: Yuxin Hu, Xiaohan Wang, Kevin Leach

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Fuzzing/Fuzz testing](<https://devfeed.tech/topics/fuzzing.md>), [test](<https://devfeed.tech/topics/test.md>)

Tags: [behavior](<https://devfeed.tech/tags/behavior.md>), [branch](<https://devfeed.tech/tags/branch.md>), [comparison](<https://devfeed.tech/tags/comparison.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [fuzzing](<https://devfeed.tech/tags/fuzzing.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-assisted](<https://devfeed.tech/tags/llm-assisted.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [reference](<https://devfeed.tech/tags/reference.md>), [repair](<https://devfeed.tech/tags/repair.md>), [symbolic-execution](<https://devfeed.tech/tags/symbolic-execution.md>)

### AI overview

SymDec uses coverage-guided symbolic execution to find inputs that distinguish an original binary from its recompiled decompilation. It replays those inputs on both programs to confirm behavioral differences, then sends structured feedback to an LLM for repair. Across 960 tasks covering x86-64 and ARM64 binaries at optimization levels O0-O3, symbolic comparison found replay-confirmed differences in 4.5% of candidates that passed all supplied tests, showing that test-suite success alone can miss discrepancies. With the same input-generation budget, symbolic execution achieved higher mean line and branch coverage than black-box fuzzing. Symbolic feedback also raised the mean single-step test-pass rate from 90.1% to 93.3%. More than half of tasks produced candidates that passed the supplied tests and had no differences detected by symbolic comparison; this indicates evidence of consistency, not proof of full behavioral equivalence.

### Source excerpt

arXiv:2611.18892v1 Announce Type: new Abstract: When source code is unavailable, decompilation provides a source-level representation of a binary. However, code recovered by conventional or LLM-assisted decompilation may not preserve the original behavior of the binary. Recent approaches evaluate re-executability using fixed test suites, but passing these suites does not ensure behavioral preservation. We present SymDec, an LLM-assisted decompiled-code repair system that uses the original binary as a behavioral reference. SymDec uses coverage-guided symbolic execution to explore the I/O-level behavior of the original and recompiled binaries, and symbolic comparison to identify inputs that expose behavioral differences. These inputs are replayed on both binaries to confirm the differences, which are returned as structured feedback to guide subsequent LLM revisions. We evaluate SymDec on 120 functions from ExeBench and DecompileBench, compiled for x86-64 and ARM64 at optimization levels O0-O3, yielding 960 binary decompilation tasks. More than half of the tasks yield candidates that pass the supplied test suite and complete symbolic comparison without detected behavioral differences. Among candidates that pass all supplied tests, symbolic comparison exposes replay-confirmed behavioral differences in 4.5% of cases, showing that test-suite re-executability can miss behavioral discrepancies. Under the same input-generation budget, symbolic execution achieves higher mean line and branch coverage than black-box fuzzing. Symbolic feedback also improves the conventional metric, increasing the mean single-step test-pass rate from 90.1% to 93.3%. These results show that symbolic comparison provides stronger evidence of behavioral consistency than test-suite re-executability alone and supplies effective feedback for repairing decompiled code.

## ProcReverser: Multi-Agent Black-Box Microarchitectural Reverse Engineering for Processor Vulnerability Analysis

DevFeed: [ProcReverser: Multi-Agent Black-Box Microarchitectural Reverse Engineering for Processor Vulnerability Analysis](<https://devfeed.tech/articles/procreverser-multi-agent-black-box-microarchitectural-reverse-engineering-for-processor-vulnerability-analysis-137619.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20827>)

Author: Guanqi Lyu, Decheng Qiu, Guangyang Zhang, Fei Zhao, Zhaojun Lu, Jiaji He, Yongqiang Lyu

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>)

Tags: [automated](<https://devfeed.tech/tags/automated.md>), [black-box](<https://devfeed.tech/tags/black-box.md>), [branch-prediction](<https://devfeed.tech/tags/branch-prediction.md>), [cache](<https://devfeed.tech/tags/cache.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [defense](<https://devfeed.tech/tags/defense.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [risc-v](<https://devfeed.tech/tags/risc-v.md>), [sifive](<https://devfeed.tech/tags/sifive.md>), [spectre](<https://devfeed.tech/tags/spectre.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

ProcReverser automates black-box microarchitectural reverse engineering to help analyze processor vulnerabilities when implementation details are unavailable. Its multi-agent system uses execution feedback to refine experiment plans and test programs, then infers properties from measurements. The proposed task suite covers speculative execution, branch prediction, memory dependence, and cache behavior relevant to Spectre-class attacks. On the XuanTie C910, it reports Pass@1/Pass@5 scores of 81.3%/93.8%; on the gem5 RISC-V O3 model, 83.0%/100.0%. On the SiFive P550, it characterizes properties covered by the suite. The abstract presents these results as evidence of potential support for vulnerability analysis and defense evaluation; it does not provide further methodological or result details.

### Source excerpt

arXiv:2611.20827v1 Announce Type: cross Abstract: When processor implementation details are unavailable, black-box microarchitectural reverse engineering helps security researchers identify hardware conditions related to security vulnerabilities. However, this process requires substantial manual effort to design experiments, develop test programs, and analyze results for each processor. We present ProcReverser, a multi-agent system that automates this process. ProcReverser uses execution feedback to improve experimental plans and test programs and identifies microarchitectural properties from measurements. We develop a comprehensive suite of property identification tasks focused on hardware mechanisms relevant to Spectre-class attacks, covering speculative execution, branch prediction, memory dependence, and cache behavior. We evaluate ProcReverser on the XuanTie C910, the gem5 RISC-V O3 CPU model, and the SiFive P550. ProcReverser achieves Pass@1 and Pass@5 scores of 81.3% and 93.8%, respectively, on C910, and 83.0% and 100.0% on gem5 O3. On the SiFive P550, ProcReverser further characterizes microarchitectural properties covered by the proposed task suite. These findings highlight the potential of automated microarchitectural reverse engineering to support processor vulnerability analysis and defense evaluation.

## Better Call Graphs: A New Dataset of Function Call Graphs for Malware Classification

DevFeed: [Better Call Graphs: A New Dataset of Function Call Graphs for Malware Classification](<https://devfeed.tech/articles/better-call-graphs-a-new-dataset-of-function-call-graphs-for-malware-classification-139227.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2512.20872>)

Author: Jakir Hossain, Jue Guo, Gurvinder Singh, Lukasz Ziarek, Ahmet Erdem Sar{\\i}y\\"uce

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Android](<https://devfeed.tech/topics/android.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>)

Tags: [android](<https://devfeed.tech/tags/android.md>), [classification](<https://devfeed.tech/tags/classification.md>), [classifiers](<https://devfeed.tech/tags/classifiers.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [features](<https://devfeed.tech/tags/features.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [malicious](<https://devfeed.tech/tags/malicious.md>), [malware](<https://devfeed.tech/tags/malware.md>), [unique](<https://devfeed.tech/tags/unique.md>)

### AI overview

Better Call Graphs (BCG) provides a dataset of large, unique function call graphs extracted from recent Android APKs to address shortcomings in existing malware-classification datasets, which the abstract describes as outdated, redundant, or unrepresentative of malware diversity. It includes benign and malicious apps from varied families and types, with graph-level features for each APK. The authors report baseline-classifier experiments demonstrating BCG's value compared with existing datasets. The supplied text does not give dataset size or detailed experimental results.

### Source excerpt

arXiv:2512.20872v3 Announce Type: replace Abstract: Function call graphs (FCGs) have emerged as a powerful abstraction for malware detection, capturing the behavioral structure of applications beyond surface-level signatures. Their utility in traditional program analysis has been well established, enabling effective classification and analysis of malicious software. In the mobile domain, especially in the Android ecosystem, FCG-based malware classification is particularly critical due to the platform's widespread adoption and the complex, component-based structure of Android apps. However, progress in this direction is hindered by the lack of large-scale, high-quality Android-specific FCG datasets. Existing datasets are often outdated, dominated by small or redundant graphs resulting from app repackaging, and fail to reflect the diversity of real-world malware. These limitations lead to overfitting and unreliable evaluation of graph-based classification methods. To address this gap, we introduce Better Call Graphs (BCG), a comprehensive dataset of large and unique FCGs extracted from recent Android application packages (APKs). BCG includes both benign and malicious samples spanning various families and types, along with graph-level features for each APK. Through extensive experiments using baseline classifiers, we demonstrate the necessity and value of BCG compared to existing datasets. BCG is publicly available at https://erdemub.github.io/BCG-dataset.

## Google found hackers using autonomous AI agents to turn security patches into exploits

DevFeed: [Google found hackers using autonomous AI agents to turn security patches into exploits](<https://devfeed.tech/articles/google-found-hackers-using-autonomous-ai-agents-to-turn-security-patches-into-exploits-136354.md>)

Original publisher: [Read original article](<https://thenewstack.io/autonomous-ai-exploit-agents/>)

Author: Amanda Caswell

Published: 2026-11-16T21:27:38Z

Content type: news

Language: en

Sources: [The New Stack](<https://devfeed.tech/sources/the-new-stack.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Security](<https://devfeed.tech/topics/security.md>), [patches](<https://devfeed.tech/topics/patches.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [threat detection](<https://devfeed.tech/topics/threat-detection.md>), [Security, Privacy and Abuse Prevention](<https://devfeed.tech/topics/security-privacy-and-abuse-prevention.md>)

Tags: [access-controls](<https://devfeed.tech/tags/access-controls.md>), [agent-harnesses](<https://devfeed.tech/tags/agent-harnesses.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [application-security](<https://devfeed.tech/tags/application-security.md>), [autonomous-ai-agents](<https://devfeed.tech/tags/autonomous-ai-agents.md>), [cairn](<https://devfeed.tech/tags/cairn.md>), [cve-2026-109618](<https://devfeed.tech/tags/cve-2026-109618.md>), [cve-2026-109619](<https://devfeed.tech/tags/cve-2026-109619.md>), [cyberstrikeai](<https://devfeed.tech/tags/cyberstrikeai.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [exploit-agent](<https://devfeed.tech/tags/exploit-agent.md>), [exploitation](<https://devfeed.tech/tags/exploitation.md>), [exploits](<https://devfeed.tech/tags/exploits.md>), [google](<https://devfeed.tech/tags/google.md>), [hackers](<https://devfeed.tech/tags/hackers.md>), [model-context-protocol-mcp](<https://devfeed.tech/tags/model-context-protocol-mcp.md>), [patches](<https://devfeed.tech/tags/patches.md>), [security](<https://devfeed.tech/tags/security.md>), [security-patches](<https://devfeed.tech/tags/security-patches.md>)

### AI overview

Google and Mandiant report attackers using multi-agent LLM frameworks to automate patch diffing, generate exploit candidates, and try them against targets. In one case, a publicly exposed server ran `exploit_agent`, which connected frontier and locally hosted models; researchers could not identify which models performed which tasks or determine whether the framework obtained the stolen data found on the server. Google also observed CAIRN in intrusions targeting U.S. aerospace organizations, while CyberStrikeAI and HexStrike AI are described as other approaches with different orchestration and tool setups. The excerpt cites rapid LLM-assisted exploit development after vulnerability disclosures as a challenge for defenders, but says the observed frameworks produce detectable network traffic and artifacts, and their hallucinations can derail attacks or expose operators. The excerpt does not establish how often generated exploits succeed.

### Source excerpt

Google's Threat Intelligence Group says attackers are using autonomous AI agents to reverse-engineer security patches, generate exploits, and attempt to The post Google found hackers using autonomous AI agents to turn security patches into exploits appeared first on The New Stack.

## REFORGE: A Method for Benchmarking LLMs' Reverse Engineering Capabilities in Decompiled Binary Function Naming

DevFeed: [REFORGE: A Method for Benchmarking LLMs' Reverse Engineering Capabilities in Decompiled Binary Function Naming](<https://devfeed.tech/articles/reforge-a-method-for-benchmarking-llms-reverse-engineering-capabilities-in-decompiled-binary-function-naming-134995.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.07738>)

Author: Nicolas Koller, Andreas U. Schmidt

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [Ground truth / benchmark quality](<https://devfeed.tech/topics/ground-truth-benchmark-quality.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [binary-analysis](<https://devfeed.tech/tags/binary-analysis.md>), [compilation](<https://devfeed.tech/tags/compilation.md>), [compiler-optimization](<https://devfeed.tech/tags/compiler-optimization.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-pl](<https://devfeed.tech/tags/cs-pl.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [llms](<https://devfeed.tech/tags/llms.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [survivorship-bias](<https://devfeed.tech/tags/survivorship-bias.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>)

### AI overview

Reforge addresses a reliability gap in benchmarking LLM function naming: compiler optimization can make binary-to-source alignment uncertain, yet existing benchmarks may report accuracy without showing how many functions are reliably evaluable. Its provenance-tracked pipeline builds function-level ground truth from C through compilation, DWARF and syntactic extraction, alignment, and decompilation, then uses an eight-gate confidence funnel and three confidence tiers to represent uncertainty. In a controlled micro-benchmark, high-confidence yield fell from 87.2% to 65.9% across optimization levels; the authors report that unpaired comparisons overstate performance decay because of survivorship bias. A proof-of-concept evaluated seven contemporary LLMs on function naming. The abstract motivates uncertainty-aware benchmarking but gives no model-specific results.

### Source excerpt

arXiv:2607.07738v3 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly applied to reverse-engineering tasks, and recent threat-intelligence reporting shows them operating inside live offensive-security workflows. Claims about their capability, however, outpace our ability to measure it. Existing benchmarks for LLM-assisted binary analysis treat the construction of function-level ground truth as a solved pre-processing step and report accuracy without disclosing how many functions were reliably evaluable. We argue that the principal obstacle to fair evaluation is not model capability but the reliability of binary-to-source alignment under compiler optimization. This paper presents Reforge, a provenance-tracked pipeline that constructs function-level ground truth from C source through compilation, DWARF and syntactic extraction, alignment, and decompilation, and that operationalizes alignment uncertainty as an eight-gate confidence funnel with three-tier stratification. On a controlled micro-benchmark, high-confidence yield falls from 87.2% to 65.9% across optimization levels, and unpaired comparisons overstate optimization-induced performance decay through survivorship bias. A proof-of-concept evaluation of seven contemporary LLMs on function naming demonstrates the validity of the concept and generally motivates an uncertainty-aware benchmarking practice.

## Reverse Engineer that USB Device in Style

DevFeed: [Reverse Engineer that USB Device in Style](<https://devfeed.tech/articles/reverse-engineer-that-usb-device-in-style-134308.md>)

Original publisher: [Read original article](<https://hackaday.com/2026/11/15/reverse-engineer-that-usb-device-in-style/>)

Author: Elliot Williams

Published: 2026-11-15T09:00:44Z

Content type: tutorial

Language: en

Sources: [Hackaday](<https://devfeed.tech/sources/hackaday.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [USB](<https://devfeed.tech/topics/usb.md>)

Tags: [computer-hacks](<https://devfeed.tech/tags/computer-hacks.md>), [drivers](<https://devfeed.tech/tags/drivers.md>), [reverse](<https://devfeed.tech/tags/reverse.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [tools](<https://devfeed.tech/tags/tools.md>), [usb](<https://devfeed.tech/tags/usb.md>)

### AI overview

usb-rea helps document USB protocols by capturing device traffic while you annotate actions in the vendor app. It timestamps traffic against those actions, then generates a report, a driver scaffold, and a Wireshark dissector. The `lookup` command checks whether a device's VID:PID is already in its dataset. The tool does not eliminate the need for proprietary software: the vendor driver or app must still run during recording. The text notes that the dataset mainly covers computer cooling and lighting peripherals, and that the author's mouse was not listed.

### Source excerpt

Proprietary device drivers, dedicated applications, and their ilk are the bane of the open-source hacker. If only we could snoop on every device out there, recording the "secret" USB transactions ...read more

## Reverse Engineering Score Perfect to Rescue its Files

DevFeed: [Reverse Engineering Score Perfect to Rescue its Files](<https://devfeed.tech/articles/reverse-engineering-score-perfect-to-rescue-its-files-133941.md>)

Original publisher: [Read original article](<https://hackaday.com/2026/11/14/reverse-engineering-score-perfect-to-rescue-its-files/>)

Author: Aaron Beckendorf

Published: 2026-11-14T12:00:47Z

Content type: news

Language: en

Sources: [Hackaday](<https://devfeed.tech/sources/hackaday.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Ghidra](<https://devfeed.tech/topics/ghidra.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>)

Tags: [binary-file](<https://devfeed.tech/tags/binary-file.md>), [data-preservation](<https://devfeed.tech/tags/data-preservation.md>), [file-format](<https://devfeed.tech/tags/file-format.md>), [file-formats](<https://devfeed.tech/tags/file-formats.md>), [ghidra](<https://devfeed.tech/tags/ghidra.md>), [musicxml](<https://devfeed.tech/tags/musicxml.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [score-perfect](<https://devfeed.tech/tags/score-perfect.md>), [software-hacks](<https://devfeed.tech/tags/software-hacks.md>)

### AI overview

Score Perfect's MusicXML export omitted chords, lyrics, and other details, so Maarten Tromp spent four years reverse-engineering its file formats to preserve his father's scores for newer software. The format changed after version four; earlier releases were Atari programs run on Windows through an emulator. After inspecting the program with Ghidra, he mapped the files by making one change at a time and comparing saved files. This revealed blocks for information such as measures and notes, plus supplemental tags. He used the findings to build a translator that exports Score Perfect files to MusicXML.

### Source excerpt

There are myriad challenges associated with data preservation - bit rot, loss of metadata, and hardware failures, among others - but one of the hardest is making sure that the ...read more

## Podcast: Election Fallout

DevFeed: [Podcast: Election Fallout](<https://devfeed.tech/articles/podcast-election-fallout-133489.md>)

Original publisher: [Read original article](<https://www.404media.co/podcast-election-fallout/>)

Author: Samantha Cole

Published: 2026-11-13T15:38:09Z

Content type: news

Language: en

Sources: [404 Media](<https://devfeed.tech/sources/404-media.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Spotify](<https://devfeed.tech/topics/spotify.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-companies](<https://devfeed.tech/tags/ai-companies.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [election](<https://devfeed.tech/tags/election.md>), [law](<https://devfeed.tech/tags/law.md>), [memes](<https://devfeed.tech/tags/memes.md>), [podcast](<https://devfeed.tech/tags/podcast.md>), [podcasts](<https://devfeed.tech/tags/podcasts.md>), [porn](<https://devfeed.tech/tags/porn.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [silicon-valley](<https://devfeed.tech/tags/silicon-valley.md>), [tech](<https://devfeed.tech/tags/tech.md>), [uk](<https://devfeed.tech/tags/uk.md>)

### AI overview

The episode says this year's election results reflect voter anger at the tech oligarchy, Silicon Valley, and data centers, and argues that right-wing memes are no longer working. It also covers AI-powered reverse engineering and the threat posed by the AI companies that enabled it. A subscribers-only segment examines the UK's new anti-porn law. The excerpt says the weekly podcast is available on Apple Podcasts, Spotify, and YouTube; paid subscribers can access the bonus episode and extended video through links from Transistor or the podcast show notes.

### Source excerpt

All those Malfoid memes meant nothing?

## GeoVision Camera Software Abused to Deploy XMRig Cryptominer

DevFeed: [GeoVision Camera Software Abused to Deploy XMRig Cryptominer](<https://devfeed.tech/articles/geovision-camera-software-abused-to-deploy-xmrig-cryptominer-133546.md>)

Original publisher: [Read original article](<https://www.huntress.com/blog/geovision-camera-software-xmrig-cryptominer>)

Author: Austin Worline; Peter Moote; Anthony Gibbs

Published: 2026-11-13T14:00:00Z

Content type: news

Language: en

Sources: [Huntress Blog](<https://devfeed.tech/sources/huntress-blog.md>)

Topics: [XMRig](<https://devfeed.tech/topics/xmrig.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>)

Tags: [1-1-1-1](<https://devfeed.tech/tags/1-1-1-1.md>), [access-control](<https://devfeed.tech/tags/access-control.md>), [attack](<https://devfeed.tech/tags/attack.md>), [attack-surface](<https://devfeed.tech/tags/attack-surface.md>), [bittorrent](<https://devfeed.tech/tags/bittorrent.md>), [bring-your-own-vulnerable-driver](<https://devfeed.tech/tags/bring-your-own-vulnerable-driver.md>), [byovd](<https://devfeed.tech/tags/byovd.md>), [camera](<https://devfeed.tech/tags/camera.md>), [cctv](<https://devfeed.tech/tags/cctv.md>), [command](<https://devfeed.tech/tags/command.md>), [commands](<https://devfeed.tech/tags/commands.md>), [cryptominer](<https://devfeed.tech/tags/cryptominer.md>), [defender](<https://devfeed.tech/tags/defender.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [exposed](<https://devfeed.tech/tags/exposed.md>), [iocs](<https://devfeed.tech/tags/iocs.md>), [net](<https://devfeed.tech/tags/net.md>), [organizations](<https://devfeed.tech/tags/organizations.md>), [software](<https://devfeed.tech/tags/software.md>), [xmrig](<https://devfeed.tech/tags/xmrig.md>)

### AI overview

Huntress observed malicious activity tied to publicly exposed GeoVision devices at 13 organizations, with activity concentrated November 8-10. GeoStartupAgentService.exe launched PowerShell scripts that probed writable directories, downloaded a driver previously abused to tamper with antivirus protections, and configured IFEO redirects targeting Microsoft Defender and Windows Security components. A later script ran a custom Golang loader, which used BitTorrent DHT and PKARR to locate its server; on some endpoints it also installed scheduled tasks for persistence. The loader ultimately downloaded XMRig. The excerpt recommends removing unnecessary public access to GeoVision interfaces, checking for suspicious shells and PowerShell from the service, reviewing the listed indicators, and isolating affected hosts while preserving evidence. It also identifies DNS, DHT traffic, and schtasks commands as detection opportunities.

### Source excerpt

Attackers abused publicly exposed GeoVision camera software at 13 organizations, tried to disable Defender, and deployed XMRig. Get the IoCs and next steps.

## LOHRE: Execution-Grounded Type Recovery via LLM-Orchestrated Hybrid Analysis

DevFeed: [LOHRE: Execution-Grounded Type Recovery via LLM-Orchestrated Hybrid Analysis](<https://devfeed.tech/articles/lohre-execution-grounded-type-recovery-via-llm-orchestrated-hybrid-analysis-131700.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14946>)

Author: Hao Jiang, Yujie Yang, Tao Li, Chenhui Cui, Liangfeng Xu, Chengjie Liu, Weiping Wen, Rubing Huang, Tsong Yueh Chen

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Remote Procedure Call (RPC)](<https://devfeed.tech/topics/rpc.md>)

Tags: [analysis](<https://devfeed.tech/tags/analysis.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [execution](<https://devfeed.tech/tags/execution.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [llm](<https://devfeed.tech/tags/llm.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [type](<https://devfeed.tech/tags/type.md>)

### AI overview

LOHRE recovers user-defined type layouts in stripped binaries by treating candidate layouts as hypotheses and testing them through paired dynamic slice emulation on baseline and mutated inputs. It compresses execution traces into an execution-depth difference and field-use summary, while a cross-function knowledge base shares compatible hypotheses and field confidence scores. Across RealType's four optimization levels, structural accuracy averaged 57.85%; on Windows RPC it was 35.6%, outperforming the evaluated static baselines. Removing dynamic validation lowered accuracy to 19.28% and 9.2%, respectively; removing the cross-function knowledge base lowered it to 41.03% and 10.0%. Using full traces instead of compressed feedback reduced accuracy and raised token use 3.7-fold and runtime 3.0-fold.

### Source excerpt

arXiv:2611.14946v1 Announce Type: new Abstract: Documenting stripped binary functions for security analysis requires recovering function interfaces and user-defined type (UDT) layouts from machine code. Existing large language model (LLM)-based methods face two limitations: they lack execution-based validation to detect inferred UDTs inconsistent with actual memory layouts, and may overlook evidence about shared UDTs across functions without direct call relationships. We present LLM-Orchestrated Hybrid Reverse Engineering (LOHRE), a framework that treats inferred UDTs as falsifiable hypotheses and tests them against program execution. LOHRE uses a hypothesize-validate-refine loop to test candidate layouts through paired dynamic slice emulation of baseline and mutated inputs. To reduce the cost of processing lengthy execution histories, the Execution-Depth Heuristic (EDH) compresses paired traces into an execution-depth difference and a summary of the tested field's observed uses. A cross-function knowledge base aggregates compatible UDT hypotheses and field-level confidence scores to guide testing and hypothesis refinement. We evaluate LOHRE on RealType at four optimization levels and a Windows remote procedure call (RPC) dataset. LOHRE achieves an average UDT Structural accuracy of 57.85% across RealType's four optimization levels and 35.6% on Windows RPC, outperforming the evaluated static baselines. Disabling dynamic validation reduces accuracy to 19.28% on RealType and 9.2% on Windows RPC. Removing the cross-function knowledge base reduces accuracy to 41.03% on RealType and 10.0% on Windows RPC. Replacing EDH's compressed feedback with full execution traces reduces recovery accuracy while increasing token use by 3.7 times and runtime by 3.0 times. These results indicate that execution-based validation, compact feedback, and cross-function evidence contribute to accurate and efficient UDT recovery.

## IoTPwnBench: A Synthetic Benchmark for IoT Firmware Security Analysis

DevFeed: [IoTPwnBench: A Synthetic Benchmark for IoT Firmware Security Analysis](<https://devfeed.tech/articles/iotpwnbench-a-synthetic-benchmark-for-iot-firmware-security-analysis-131711.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15316>)

Author: Luca Ferrari, Marco Alecci, Luca Verderame, Jacques Klein, Francesco Pagano

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [Software Testing](<https://devfeed.tech/topics/software-testing.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [firmware-security](<https://devfeed.tech/tags/firmware-security.md>), [iot](<https://devfeed.tech/tags/iot.md>), [security](<https://devfeed.tech/tags/security.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>)

### AI overview

IoTPwnBench provides reproducible IoT firmware targets with known weaknesses for evaluating security analysis tools. It includes 716 vulnerable images spanning 86 scenarios, 28 CWE classes, and 11 device profiles. Each runs a plausible device application with simulated sensors and actuators, exposes one weakness, and has a documented exploitation payload and observable signal. A corrected twin preserves the device and service while removing that weakness, enabling paired assessment of detection and false positives. The benchmark produces kernel and filesystem binaries for static inspection and QEMU dynamic testing; separate device profiles and weakness modules allow scenarios to be reused across contexts. The abstract reports manual validation of vulnerable targets and a bit-for-bit reproducible open source build workflow.

### Source excerpt

arXiv:2611.15316v1 Announce Type: cross Abstract: Building and evaluating security tools for IoT firmware requires targets whose weaknesses are known in ad- vance and can be reproduced in a running device. Two problems remain: (1) firmware collected from real-world products provides realistic software, but hardware dependencies and undocumented defects make it difficult to establish what a security tool should be expected to find; and (2) existing benchmarks typically collect sets of vulnerable or misconfigured firmware images without providing purpose-built vulnerable services with explicit weaknesses that a security analysis tool can target. We present IoTPwnBench, a synthetic, modular benchmark that brings these requirements together. IoTPwnBench contains 716 vulnerable firmware images covering 86 weakness scenarios and 28 CWE classes across 11 device profiles, including industrial controllers, health monitors, and household devices. Each vulnerable image runs a plausible device application with simulated sensors and actuators that generate live readings over its native interfaces while exposing one target weakness mapped to a primary CWE. A documented exploitation payload and observable signal establish the expected behavior independently of the evaluated tool. Every vulnerable image is paired with a corrected twin that preserves the device and service while removing that weakness. Researchers can therefore compare the same target before and after correction to assess detection and false positives for the target weakness. The benchmark produces kernel and filesystem binaries for static inspection and dynamic testing under QEMU. Device profiles and weakness modules remain separate, allowing the same scenario to be reused across device contexts. Vulnerable targets were validated manually, and the open source build workflow regenerates the firmware bit-for-bit.

## Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial Training

DevFeed: [Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial Training](<https://devfeed.tech/articles/representational-simplicity-and-circuit-size-dissociate-in-a-threshold-dependent-way-a-controlled-test-via-adversarial-training-131975.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2609.35890>)

Author: Adam Elimadi

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Adversarial attacks](<https://devfeed.tech/topics/adversarial-attacks.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [adversarial](<https://devfeed.tech/tags/adversarial.md>), [apply](<https://devfeed.tech/tags/apply.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attribution](<https://devfeed.tech/tags/attribution.md>), [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [but](<https://devfeed.tech/tags/but.md>), [checkpoint](<https://devfeed.tech/tags/checkpoint.md>), [circuit](<https://devfeed.tech/tags/circuit.md>), [circuits](<https://devfeed.tech/tags/circuits.md>), [compare](<https://devfeed.tech/tags/compare.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [easier](<https://devfeed.tech/tags/easier.md>), [engagement](<https://devfeed.tech/tags/engagement.md>), [simplicity](<https://devfeed.tech/tags/simplicity.md>), [size](<https://devfeed.tech/tags/size.md>), [test](<https://devfeed.tech/tags/test.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

Using matched standard and adversarial continual training from the same GPT-2 Small checkpoint, the study compares SAE decomposability, SAE feature engagement in task attribution, and faithful circuit size, after requiring retained indirect object identification competence and independent robustness verification. The robust model is more SAE-decomposable and uses fewer SAE features in attribution, but these properties do not consistently predict smaller circuits: on competence-matched IOI, standard training leads or ties below 85% faithfulness, while the robust model needs substantially fewer circuit edges at 90% and 95%. This high-faithfulness advantage is reported across all qualified models in a seven-strength sweep and a second corpus; representational trends also hold there.

### Source excerpt

arXiv:2609.35890v2 Announce Type: replace Abstract: Sparse-autoencoder decomposability and concentrated feature attribution are increasingly treated as evidence that a model's computation is easier to reverse-engineer. Whether this representational and attributional cleanliness actually predicts a smaller or more tractable causal circuit remains an open question. We test this directly using adversarial training as a controlled instrument: it reliably reshapes internal representations, but this alone does not constitute a test of circuit size. We investigate this question through reverse-engineering complexity: the causal structure required to recover a model's behavior at a fixed level of faithfulness. To our knowledge, this is the first controlled empirical test of whether representational or attributional simplicity translates into causal simplicity at the circuit level. Starting from the same pretrained GPT-2 Small checkpoint, we apply matched standard and adversarial continual training, requiring both conditions to retain competence on indirect object identification and pass independent robustness verification before comparing mechanisms. We then compare the models along three complementary axes: sparse-autoencoder decomposability, SAE feature engagement in task attribution, and the size of faithful circuits recovered from the raw computational graph. The robust model is more SAE-decomposable and engages fewer SAE features in task attribution. Circuit size is regime-dependent: on competence-matched IOI, standard leads or ties below 85% faithfulness, but robust needs substantially fewer edges at high faithfulness (90%, 95%). The high-faithfulness advantage holds for every qualified robust model in a seven-point adversarial-strength sweep and for a second corpus, as do the representational trends.

## ASAP: Assembly-Source Aligned Pseudocode Refinement For Binary Decompilation

DevFeed: [ASAP: Assembly-Source Aligned Pseudocode Refinement For Binary Decompilation](<https://devfeed.tech/articles/asap-assembly-source-aligned-pseudocode-refinement-for-binary-decompilation-131679.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.06900>)

Author: Yujian Zhuang, Dehong Gao, Qichao Zhang, Qijing Lai, Jiaxin Wang, Libin Yang, Xiaoyan Cai

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>), [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>), [Assembly](<https://devfeed.tech/topics/assembly.md>), [code analysis](<https://devfeed.tech/topics/code-analysis.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>)

Tags: [alignment](<https://devfeed.tech/tags/alignment.md>), [assembly](<https://devfeed.tech/tags/assembly.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [binaries](<https://devfeed.tech/tags/binaries.md>), [c](<https://devfeed.tech/tags/c.md>), [compiler](<https://devfeed.tech/tags/compiler.md>), [compiler-optimization](<https://devfeed.tech/tags/compiler-optimization.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-pl](<https://devfeed.tech/tags/cs-pl.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [decompiler](<https://devfeed.tech/tags/decompiler.md>), [errors](<https://devfeed.tech/tags/errors.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models-llms](<https://devfeed.tech/tags/large-language-models-llms.md>), [pseudocode](<https://devfeed.tech/tags/pseudocode.md>), [source](<https://devfeed.tech/tags/source.md>)

### AI overview

ASAP refines decompiler-produced pseudocode by conditioning a decompilation LLM on both that pseudocode and assembly features aligned with source code. It learns the alignment from paired source and binary functions, compresses assembly features into a fixed set of tokens, and uses stochastic pseudocode masking plus a relative assembly-advantage loss to reduce reliance on potentially lossy pseudocode alone. Across two benchmarks and multiple compiler optimization levels, it raises average re-execution rates from 64.7% to 71.9% and recompilation rates from 91.6% to 96.6% versus the strongest baseline.

### Source excerpt

arXiv:2610.06900v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used in binary decompilation to refine the C-like pseudocode produced by traditional rule-based decompilers. While this pseudocode is useful, it is a heuristic and lossy abstraction rather than a faithful copy of the source code. It often contains decompiler errors, especially for aggressively optimized binaries where critical low-level details are obscured. We present ASAP, an Assembly-Source Aligned Pseudocode refinement framework for binary decompilation. ASAP learns source-aligned assembly representations from paired source and binary functions using joint function-level and snippet-level contrastive alignment. A Q-Former then compresses chunk-level assembly features into a fixed number of assembly tokens. These tokens and the decompiler-produced pseudocode jointly condition the decompilation LLM. During refinement, we use stochastic pseudocode masking and a relative assembly-advantage loss to discourage the model from relying solely on pseudocode. On two decompilation benchmarks across multiple compiler optimization levels, ASAP improves the average re-execution rate from 64.7% to 71.9% and the average recompilation rate from 91.6% to 96.6% compared with the strongest baseline. Our approach offers both a new perspective and a practical solution to binary decompilation.

## Hacking Meta's Smart Glasses

DevFeed: [Hacking Meta's Smart Glasses](<https://devfeed.tech/articles/hacking-meta-s-smart-glasses-131177.md>)

Original publisher: [Read original article](<https://hackaday.com/2026/11/12/hacking-metas-smart-glasses/>)

Author: Jenny List

Published: 2026-11-13T03:00:45Z

Content type: tutorial

Language: en

Sources: [Hackaday](<https://devfeed.tech/sources/hackaday.md>)

Topics: [Hacking](<https://devfeed.tech/topics/hacking.md>), [Glasses & AR Tech](<https://devfeed.tech/topics/glasses-ar-tech.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>)

Tags: [apk](<https://devfeed.tech/tags/apk.md>), [app](<https://devfeed.tech/tags/app.md>), [application](<https://devfeed.tech/tags/application.md>), [applications](<https://devfeed.tech/tags/applications.md>), [bounty](<https://devfeed.tech/tags/bounty.md>), [bug](<https://devfeed.tech/tags/bug.md>), [bug-bounty](<https://devfeed.tech/tags/bug-bounty.md>), [bug-bounty-program](<https://devfeed.tech/tags/bug-bounty-program.md>), [code](<https://devfeed.tech/tags/code.md>), [controversy](<https://devfeed.tech/tags/controversy.md>), [decompilation](<https://devfeed.tech/tags/decompilation.md>), [developer](<https://devfeed.tech/tags/developer.md>), [device](<https://devfeed.tech/tags/device.md>), [firmware](<https://devfeed.tech/tags/firmware.md>), [glasses](<https://devfeed.tech/tags/glasses.md>), [hacking](<https://devfeed.tech/tags/hacking.md>), [meta](<https://devfeed.tech/tags/meta.md>), [ray-ban](<https://devfeed.tech/tags/ray-ban.md>), [rayban](<https://devfeed.tech/tags/rayban.md>), [security-hacks](<https://devfeed.tech/tags/security-hacks.md>), [smart-glasses](<https://devfeed.tech/tags/smart-glasses.md>), [surveillance](<https://devfeed.tech/tags/surveillance.md>)

### AI overview

Unable to probe the glasses' hardware, researcher p0pcycle intercepted a software update and decompiled the companion app to obtain firmware for analysis. An early firmware version trusted self-signed developer apps loaded as APKs, allowing arbitrary code; after experimentation, she exposed a network listener. She reported the findings through Meta's bug bounty program. The excerpt says these steps no longer work on newer firmware.

### Source excerpt

Meta's Ray-Ban smart glasses have attracted plenty of controversy in the relatively short time they have been available, given their potential for covert and unvanted surveillance. But given the chance ...read more

## Introducing ProtoSpector: A Burp Suite Extension for Analyzing, Tampering With, and Testing Protobuf and gRPC Traffic

DevFeed: [Introducing ProtoSpector: A Burp Suite Extension for Analyzing, Tampering With, and Testing Protobuf and gRPC Traffic](<https://devfeed.tech/articles/introducing-protospector-a-burp-suite-extension-for-analyzing-tampering-with-and-testing-protobuf-and-grpc-traffic-131568.md>)

Original publisher: [Read original article](<https://blog.doyensec.com/2026/11/12/protospector.html>)

Published: 2026-11-11T23:00:00Z

Content type: release

Language: en

Sources: [Doyensec](<https://devfeed.tech/sources/doyensec.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [gRPC](<https://devfeed.tech/topics/grpc.md>), [Reverse Engineering](<https://devfeed.tech/topics/reverse-engineering.md>)

Tags: [analyze](<https://devfeed.tech/tags/analyze.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [automation](<https://devfeed.tech/tags/automation.md>), [burp-suite](<https://devfeed.tech/tags/burp-suite.md>), [doyensec](<https://devfeed.tech/tags/doyensec.md>), [doyensecurity](<https://devfeed.tech/tags/doyensecurity.md>), [exploit](<https://devfeed.tech/tags/exploit.md>), [extension](<https://devfeed.tech/tags/extension.md>), [grpc](<https://devfeed.tech/tags/grpc.md>), [hacking](<https://devfeed.tech/tags/hacking.md>), [infosec](<https://devfeed.tech/tags/infosec.md>), [offensive](<https://devfeed.tech/tags/offensive.md>), [pentest](<https://devfeed.tech/tags/pentest.md>), [pentester](<https://devfeed.tech/tags/pentester.md>), [protobuf](<https://devfeed.tech/tags/protobuf.md>), [reverse-engineering](<https://devfeed.tech/tags/reverse-engineering.md>), [security](<https://devfeed.tech/tags/security.md>), [testing](<https://devfeed.tech/tags/testing.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

ProtoSpector is a Burp Suite extension for decoding, editing, and testing supported Protobuf and gRPC traffic. With preloaded .proto schemas compiled into .pb descriptors, it matches and decodes messages; without a schema, it infers a structure that users can edit for outgoing requests, including in Repeater. Its field table supports inline name and type changes and previews of alternate interpretations, though previews can fail when data is incompatible or too short. A global tab manages method descriptors, and loaded gRPC methods can be sent to Repeater. Current support is limited to listed gRPC-Web and raw Protobuf content types; standalone descriptor persistence is limited and those descriptors cannot be edited. The article also reports intermittent failures in black-box type previews and plans broader content-type support and preview refinements.

### Source excerpt

Testing protocol buffers over HTTP can be a notoriously tedious process. Today, we're excited to announce ProtoSpector, a Burp Suite extension designed to seamlessly decode and encode Protocol Buffer messages on the fly. By eliminating the need to constantly switch to external tools or instrumented browser applications, ProtoSpector dramatically accelerates your testing workflow. Whether you want to import an existing Protobuf schema or reverse-engineer descriptors from scratch, this extension gives you the full capability to analyze, tamper with, and test gRPC traffic directly within Burp. Features 🔄 Intelligent Message Decoding & Encoding The extension automatically detects supported Protobuf content types and adds a ProtoSpector tab to the request view. Two testing approaches are supported: White-box Testing: Automatically matches and decodes messages using pre-loaded .proto schemas (.pb descriptor files compiled via protoc). Black-box Testing: If no schema is found, it automatically infers the underlying structure, allowing you to view and fully edit the decoded message for outgoing requests (e.g., inside the Repeater tool). 🛠 Dynamic Signature Editing For black-box testing scenarios, the ProtoSpector tab includes an interactive fields table that displays inferred message descriptors. Inline Editing: Easily update field names or data types on the fly by clicking directly into the table cells. Type Previewing: Instantly preview how any field renders when interpreted as a different data type, using the built-in selector. If a preview fails, it is typically due to an incompatibility, such as having insufficient bytes to parse the data into a larger type. Toggleable Interface: The signature editing table can be hidden or shown at any time via the Settings tab. 🗺 Advanced Schema & Method Management Global Extension Tab: Deeply inspect, manage, and customize the signature of every loaded method descriptor across your entire session. gRPC Automation: Once a .pb file

[Next page](<https://devfeed.tech/topics/reverse-engineering.md?cursor=WyIyMDI2LTExLTExVDIzOjAwOjAwKzAwOjAwIiwgIjA5YzE3OTI5LWRlODQtNGM1OC05ZmI5LWM5ODYxNDY4ZmM0MSJd>)