“reconstruction”的搜索结果

使用稀疏自编码器将语言模型内部的激活分解为少量特征，研究模型学到的概念。涵盖预训练SAE分析、自定义训练、重建质量评估，以及利用特征调整输出的实验。

• 使用预训练SAE研究特征
• 针对特定模型层训练SAE
• 比较不同输入文本的特征响应

面向开发与研究，指导使用RuView整合多个WiFi观测点的信号，研究位置估计、三维空间占用重建、长期运动趋势变化，以及针对可疑信号的防御。

• 整合多个节点的WiFi观测
• 通过RF层析成像重建空间
• 研究长期运动趋势变化

分阶段调查GitHub项目的代码、开发历史及相关信息，将技术架构、竞品差异和活动情况整理为附有来源与图表的Markdown报告。

• 调查开源项目的技术架构
• 按时间梳理开发里程碑
• 比较竞争项目

Rebuild slide images, screenshots, scanned PDFs, or image-only presentations as editable PowerPoint files. Reconstruct text and simple shapes as separate objects, retain complex visuals as image assets where needed, and check rendered slides against their sources.

中文简介正在准备中。当前显示英语简介。

• Recovering editable slides from screenshots
• Converting scanned presentation PDFs into PowerPoint
• Rebuilding diagrams with editable shapes and connectors

Reconstruct how an AI coding agent produced a feature, pull request, or report using local ax records. Connect sessions, commits, skills, and tool activity into an evidence-based sequence with key decisions, checks, and a repeatable recipe.

中文简介正在准备中。当前显示英语简介。

• Tracing how a shipped feature was built
• Reconstructing development from a commit
• Finding the workflow behind work done on a known date

Analyze packet captures in a CTF sandbox by rebuilding communication sessions and decoding message contents. Recover transferred files and connect packet order, timing, and reconstructed streams to observed host or malware behavior.

中文简介正在准备中。当前显示英语简介。

• Rebuild TCP or UDP communication sessions
• Decode HTTP, DNS, or custom protocol messages
• Recover files transferred over the network

Analyze a custom protocol in a CTF sandbox and reproduce an accepted exchange. Recover message boundaries, checksums, and session state, then build a small replay program that preserves the required message order.

中文简介正在准备中。当前显示英语简介。

• Recover message boundaries in an unknown protocol
• Reconstruct handshake steps and session state
• Build a replay program for a captured exchange

Build an evidence-backed incident timeline in a CTF sandbox by correlating logs, network captures, and other artifacts. Trace attacker actions across sources, distinguish confirmed events from inferred gaps, and preserve the sequence that supports the conclusion.

中文简介正在准备中。当前显示英语简介。

• Combine logs and network captures into an incident timeline
• Trace attacker actions from initial access to persistence
• Separate confirmed event order from inferred gaps

Trace uploaded files through extraction, conversion, parsing, and reconstruction of saved objects in a CTF sandbox, an isolated competition environment. Preserve original and derived files, distinguish visible validation from backend behavior, and document the smallest reproducible processing chain.

中文简介正在准备中。当前显示英语简介。

• Trace an uploaded file to its backend effect
• Investigate archive extraction and preview generation
• Compare visible validation with backend file handling

Investigate memory dumps, disk records, and captured network traffic while preserving evidence. The skill guides timeline reconstruction and examination of system traces to identify suspicious activity and produce findings that others can review.

中文简介正在准备中。当前显示英语简介。

• Preserve evidence before starting an incident investigation
• Inspect processes and connections in memory dumps
• Build timelines from event logs and execution traces

Create a structured review after a production incident, focusing on systems and processes without blaming individuals. Reconstruct events, quantify impact, investigate causes, and document specific follow-up actions with named owners and deadlines.

中文简介正在准备中。当前显示英语简介。

• Write a review after a production outage
• Reconstruct an incident timeline from records
• Separate root causes from contributing factors

Check training data and machine learning models for malicious changes, including backdoors that cause unexpected behavior on specific inputs. Combine data quality checks, model analysis, and file integrity verification, then document findings and retest after remediation.

中文简介正在准备中。当前显示英语简介。

• Screening third-party data before training
• Checking downloaded model weights before deployment
• Investigating unexpected behavior on specific inputs

Build a shared incident timeline in Timesketch from endpoint, server, and cloud logs. Search events, add tags and investigator notes, reconstruct the attack sequence, and organize findings into a story for incident reporting.

中文简介正在准备中。当前显示英语简介。

• Reconstructing the sequence of an incident
• Reviewing evidence with multiple analysts
• Combining Plaso output and other logs

Analyze Windows Amcache records to investigate applications, deleted executable traces, and drivers. Extract file details, compare SHA-1 hashes with malware databases, and combine records into a timeline to support incident investigations.

中文简介正在准备中。当前显示英语简介。

• Investigating traces of deleted executables
• Checking software installed during an incident
• Comparing file hashes with malware databases

Investigate security incidents in Splunk by connecting authentication, endpoint, and network logs. Use SPL, Splunk’s search language, to identify suspicious activity, reconstruct event timelines, document findings, and develop detection rules.

中文简介正在准备中。当前显示英语简介。

• Investigating suspicious logins across multiple log sources
• Reconstructing an incident timeline
• Tracing suspicious processes and outbound traffic

Investigate Linux intrusions using auditd records. Configure audit rules, search events with ausearch, summarize activity with aureport, reconstruct a timeline of suspicious actions, and forward logs to a central security monitoring system.

中文简介正在准备中。当前显示英语简介。

• Investigating unauthorized access or privilege escalation
• Checking changes to critical files and SSH keys
• Reconstructing an incident timeline

Use SAELens to train and analyze sparse autoencoders, which separate neural network activity into features that may represent recognizable concepts. Explore pretrained features, evaluate reconstruction quality, and investigate how individual features influence language model outputs.

中文简介正在准备中。当前显示英语简介。

• Exploring features with a pretrained SAE
• Training and evaluating a custom SAE
• Studying overlapping concepts in model activity

Organize responses to service outages and degraded performance, from initial assessment through recovery review. Classify severity, reconstruct event timelines, draft stakeholder updates, and create response guides and follow-up actions based on incident findings.

中文简介正在准备中。当前显示英语简介。

• Classify outage severity and plan initial response
• Reconstruct an incident timeline from scattered events
• Draft incident updates for customers and internal teams

Build and test threat detection rules in Google SecOps using YARA-L 2.0, its rule language. The skill also checks existing coverage with simulated attack events, proposes rules for confirmed gaps, and deploys approved rules.

中文简介正在准备中。当前显示英语简介。

• Creating detection rules from known attack behavior
• Tuning existing rules to reduce false positives
• Checking detection gaps against threat reports and CVE disclosures

Build a Three.js 3D model from an object or character reference image using code. The workflow develops shapes and materials in stages, compares rendered views with the reference, and guides corrections through quality checks.

中文简介正在准备中。当前显示英语简介。

• Rebuilding a photographed object as a browser-based 3D prop
• Creating a stylized character from a reference image
• Planning model parts, materials, and animation pivots

Assess LoRaWAN and sub-GHz wireless devices by capturing signals, examining packet formats, and checking key reuse and replay weaknesses, where recorded messages are resent. It covers IoT networks, fixed-code remotes, and proprietary radio protocols.

中文简介正在准备中。当前显示英语简介。

• Reviewing LoRaWAN keys and frame counters
• Testing fixed-code remotes for replay weaknesses
• Analyzing proprietary wireless packet formats

Assess GraphQL APIs during penetration tests, from finding endpoints and mapping data structures to checking access controls, query limits, subscriptions, and uploads. Includes example queries, guidance for testing tools, and defensive controls for interpreting findings.

中文简介正在准备中。当前显示英语简介。

• Map GraphQL endpoints and data structures
• Check access controls across users and roles
• Review query limits and subscription security

Plan authorized security tests involving phone calls, text messages, physical access, and USB devices. Helps develop scenarios from public information, document interactions, and assess how staff verification procedures and physical controls respond.

中文简介正在准备中。当前显示英语简介。

• Planning authorized phone and SMS security tests
• Reviewing visitor and badge verification procedures
• Designing USB device security exercises

Investigate exposed source maps, configuration files, Git history, and API specifications on websites. The skill guides source reconstruction and artifact inspection to identify leaked secrets, internal routes, and build details during the initial stages of security research.

中文简介正在准备中。当前显示英语简介。

• Checking source maps for exposed source code
• Inspecting public configuration files and Git history
• Mapping API routes from Swagger and OpenAPI specifications