🏠 Startseite
Vergleiche
📊 Alle Benchmarks 🦖 Dinosaurier v1 🦖 Dinosaurier v2 ✅ To-Do-Listen-Apps 🎨 Kreative freie Seiten 🎯 FSACB - Ultimatives Showcase 🌍 Übersetzungs-Benchmark
Modelle
🏆 Top 10 Modelle 🆓 Kostenlose Modelle 📋 Alle Modelle ⚙️ Kilo Code
Ressourcen
💬 Prompt-Bibliothek 📖 KI-Glossar 🔗 Nützliche Links

KI-Glossar

Das vollständige Wörterbuch der Künstlichen Intelligenz

162
Kategorien
2.032
Unterkategorien
23.060
Begriffe
📖
Begriffe

Contention

Mechanism designed to restrict or guide the output of an LLM to prevent the generation of unwanted, dangerous, or out-of-scope content.

📖
Begriffe

Prompt Guardrails

Set of rules and filters applied upstream to user input to detect and block malicious, inappropriate requests, or those attempting to bypass the model's security policies.

📖
Begriffe

Output Filtering

Post-generation security mechanism that analyzes the LLM's response to identify and remove prohibited content before it is presented to the user.

📖
Begriffe

Jailbreaking

Set of reverse engineering techniques aimed at bypassing an LLM's contention and security mechanisms to force it to produce normally prohibited responses.

📖
Begriffe

Safety Layer

Distinct software component, often a classification model, that intercepts LLM inputs and outputs to evaluate their compliance with security policies.

📖
Begriffe

Decoding Alignment

Strategy of modifying the decoding process (e.g., beam search, sampling) to penalize the generation of tokens or token sequences associated with unsafe content.

📖
Begriffe

Self-Critique

Ability of an LLM to evaluate its own generated response against a set of predefined criteria (coherence, safety, accuracy) and revise it if necessary.

📖
Begriffe

Adversarial Suffix

Learned character sequence added to the end of a prompt to manipulate the LLM's internal behavior and force a specific output, often used in jailbreaking attacks.

📖
Begriffe

Preference Modeling

Process of creating a reward model that learns human preferences from pairwise response comparisons, essential for RLHF.

📖
Begriffe

Refusal Training

Specialized training phase where the LLM learns to identify inappropriate requests and generate polite and informative refusal responses instead of attempting to answer.

📖
Begriffe

Harmlessness Classification

Binary classification task to determine if an LLM output is 'harmless' or 'harmful', often implemented as a safety filter.

📖
Begriffe

Sycophancy Mitigation

Set of techniques aimed at reducing an LLM's tendency to agree with incorrect user premises to please them, an undesirable behavior that compromises truthfulness.

📖
Begriffe

Model Steering

Technique for dynamically adjusting an LLM's behavior during inference, often by modifying logits, to guide generation towards a desired and safe response space.

🔍

Keine Ergebnisse gefunden