PAGEFIND · FULL-TEXT SEARCH
WEBAI ATLAS · LOCAL DISCOVERY
All of WebAI, one search
Search the sections, explainers, tools, sources and terms included in this preview.
Search resources
230 resultsContent and resource directory
Filter by type to find tools, references and definitions.
article
Understanding the Generative Web
Creation, adaptation and publication with generative models.
Open ↗article
WebMCP: an interface for Web agents
Expose structured capabilities from a Web application.
Open ↗article
SEO, AEO and GEO: key differences
Visibility in conventional search and answer interfaces.
Open ↗article
Knowledge graphs and JSON-LD
Structure links between concepts, sources and technologies.
Open ↗article
Accessibility and AI: design for everyone
Interfaces usable by keyboard, readable and understandable.
Open ↗article
WebXR and generative experiences
Potential intersection of immersive Web and generative AI.
Open ↗article
Prompt injection and agent security
Keep authorized instructions separate from untrusted data.
Open ↗article
Visibility in generative search
Discover and evaluate content inclusion in AI answer interfaces.
Open ↗article
Control search and AI crawlers
Understand the distinct roles of indexing, search and training crawlers.
Open ↗article
llms.txt: proposal and limits
Avoid treating an optional convention as a universal ranking standard.
Open ↗article
Build websites for humans and agents
Make interfaces easier to interpret without degrading human experience.
Open ↗article
MCP: versions, capabilities and security
Understand protocol evolution without conflating adoption with universal maturity.
Open ↗article
A2A: discovering and describing agents
Understand agent cards without oversharing internal information.
Open ↗article
WebMCP: JavaScript tools exposed by a page
Study the experimental API and the exact status of the community draft.
Open ↗article
Indirect prompt injection through Web content
Separate authorized instructions from untrusted retrieved Web content.
Open ↗article
Assess risks of the generative Web
Structure risk assessment rather than relying on promises.
Open ↗article
Built-in browser AI: capabilities and limits
Assess translation, summarization and generation APIs offered by some browsers.
Open ↗article
WebNN and neural acceleration
Understand a Web API designed for inference on user devices.
Open ↗article
WebGPU for computation and Web models
Separate GPU computation from model-level AI logic.
Open ↗article
Rigorous evaluation of RAG systems
Identify errors at each stage of retrieval-augmented generation.
Open ↗article
Structured data, JSON-LD and discoverability
Understand relationships between semantics, metadata and AI systems.
Open ↗article
C2PA provenance and independent verification
Avoid conflating signatures, attribution and factual truth.
Open ↗article
EU transparency for AI-generated content
Read transparency requirements without generalizing beyond applicable law.
Open ↗article
Accessible and understandable AI interfaces
Apply WCAG to dynamic content and AI interactions.
Open ↗article
Right-sized AI and sustainability on the Web
Choose model architecture proportionate to tasks and devices.
Open ↗article
Teaching critical AI literacy for the synthetic Web
Move from passive acceptance of answers to evidence-led inquiry.
Open ↗source
ReAct — Reasoning and Acting (2022)
Study of combining reasoning traces and actions involving sources or environments. Useful for explaining tool-using agents without establishing general robustness.
Open ↗source
Toolformer — Language Models Using Tools (2023)
NeurIPS work on learning to call external tools from language models. Experiments focus on selected APIs and tasks.
Open ↗source
Lost in the Middle — Long Context (2024)
Published TACL study showing that position of relevant information in long contexts can affect model use. Important for RAG design and testing retrieved passages.
Open ↗source
WebArena — Realistic Web Agents (2023)
Self-hostable research environment with realistic Web tasks for evaluating agents’ ability to complete actions. Findings depend on benchmark sites and scenarios.
Open ↗source
VisualWebArena — Visual Web Tasks (2024)
Web benchmark where visual perception complements text interpretation, supporting evaluation of agents navigating human-facing interfaces.
Open ↗source
OSWorld — Computer Use Benchmark (2024)
Benchmark of computer tasks spanning applications and system states, illustrating the difficulty of reliably completing real operations.
Open ↗source
RAGAS — RAG Evaluation (2023)
Research on evaluating RAG systems along dimensions including passage retrieval and generated-answer faithfulness. Automated metrics require validation.
Open ↗source
GEO — Generative Engine Optimization (2023)
Study of strategies and visibility metrics in generative search engines. Results apply to experimental settings, not guaranteed universal rankings.
Open ↗source
SWE-bench — Software Agents (2024)
ICLR benchmark of code repair tasks derived from real repository issues, using executable checks rather than prose evaluation alone.
Open ↗source
AgentDojo — Prompt Injection Evaluation
Research environment for assessing prompt injections and defenses in tool-using agents. It separates simulated attacks from observed incidents.
Open ↗terme
Agentic Web
A Web also used by agents that can discover tools and carry out authorized tasks.
Open ↗terme
Large language model
A statistical model trained on large quantities of text and sometimes other modalities.
Open ↗terme
Retrieval-augmented generation
A method combining document retrieval with answer generation.
Open ↗terme
Hybrid search
Combination of multiple retrieval approaches, such as keyword and vector search.
Open ↗terme
Prompt injection
Attempt to redirect a system through instructions embedded in untrusted data.
Open ↗terme
Provenance
Information describing a piece of content’s origin and certain transformations.
Open ↗terme
Content Credentials
Provenance information associated with certain images and other media.
Open ↗terme
Model Context Protocol
Open protocol for connecting AI applications to tools and resources.
Open ↗terme
Agent2Agent
Protocol for communication and collaboration between independent agent systems.
Open ↗terme
Search engine optimization
Practices to improve site accessibility and interpretation by search engines.
Open ↗terme
Answer engine optimization
Editorial term for practices aimed at visibility in answer engines.
Open ↗terme
Generative engine optimization
Research and practice concerning visibility in generative responses.
Open ↗No results. Try another term.
Local filtering of the current collection. Semantic search and the full scientific corpus are planned after editorial review.