Zum Hauptinhalt springen
tsecurity.de LIVE
Echtzeit-Radar & Feeds
Alle RSS Feeds
👥 Community & Social
Windows Tipps & SecurityNighthawk M7 Pro im Test: Flexibler, aber teurer 5G-Router(21.09.2026 um 10:30 Uhr)
Sichere ProgrammierungNeue Gmail-Funktion: So sparst du jetzt Zeit bei Einmalcodes(21.09.2026 um 10:00 Uhr)
Sichere ProgrammierungYour GIF exporter is fine — the container is the problem(21.09.2026 um 10:01 Uhr)
Sichere ProgrammierungCSS, Motion, or GSAP? I Choose by Who Owns the Animation(21.09.2026 um 10:12 Uhr)
Windows Tipps & SecurityNighthawk M7 Pro im Test: Flexibler, aber teurer 5G-Router(21.09.2026 um 10:30 Uhr)
Sichere ProgrammierungNeue Gmail-Funktion: So sparst du jetzt Zeit bei Einmalcodes(21.09.2026 um 10:00 Uhr)
Sichere ProgrammierungYour GIF exporter is fine — the container is the problem(21.09.2026 um 10:01 Uhr)
Sichere ProgrammierungCSS, Motion, or GSAP? I Choose by Who Owns the Animation(21.09.2026 um 10:12 Uhr)
Intelligence View
⚡ tsecurity.de Intelligence

0% vs 50%: Making a RAG Agent Refuse to Hallucinate

0 % vs 50 %: making a RAG agent refuse to hallucinate 2026-05-31 · LLM / RAG A retrieval-augmented agent is only as trustworthy as its behaviour on questions whose answer isn't in the corpus. The failure mode is quiet: instead of saying …

0
↗ Quelle (dev.to)
Reagiere als Erste:r — dein Feedback zählt!




0 % vs 50 %: making a RAG agent refuse to hallucinate



2026-05-31 · LLM / RAG



A retrieval-augmented agent is only as trustworthy as its behaviour on questions whose answer

isn't in the corpus. The failure mode is quiet: instead of saying "I don't know," the model

invents a confident, well-formed, wrong answer. This post shows a single guardrail that takes

that from common to never — and, crucially, measures it.



Reference architecture:

nim-agent-blueprint — agentic RAG on

the NVIDIA NIM stack with a built-in eval harness.






The ablation



The agent loop is plan → retrieve → generate → validate. The interesting variable is the

generation prompt's contract with the retrieved context:




















Configuration Out-of-corpus hallucination rate
Generate freely from context ~50 %
Guarded prompt (answer only from context; otherwise abstain) 0 %


Same model, same retriever, same questions. The only change is a prompt that makes "I can't

answer that from the provided sources" a first-class, rewarded output — plus a validate

step that checks the answer is grounded in retrieved spans before returning it. On in-corpus

questions, retrieval recall@3 stayed at 94–100 %, so the guardrail buys safety without

costing coverage.






Why "just prompt better" isn't the lesson



The lesson isn't the prompt — it's that the difference between 50 % and 0 % is invisible

without an eval harness
. A demo that only asks in-corpus questions looks perfect in both

configurations. You only see the 50 % when you deliberately ask things the corpus can't

answer and score groundedness. So the blueprint ships with:





  • retrieval hit-rate (is the answer even retrievable?),


  • answer groundedness via LLM-as-judge (is the answer supported by what was retrieved?),


  • latency, and OpenTelemetry traces per agent step.



That's the difference between "it works on my five questions" and "here is the number a

partner can hold me to."






Takeaway



For enterprise RAG, abstention is a feature, not a failure. Make "I don't know" a rewarded

output, validate groundedness before returning, and measure the out-of-corpus rate — it's

the number that separates a demo from something you'd put in front of a customer.



→ Runnable blueprint + eval harness:

github.com/waynehacking8/nim-agent-blueprint

Ähnliche Beiträge
🔍 Verwandte News

Auch interessante Nachrichten 0% vs 50%: Making a RAG Agent Refuse to Hallucinate

Thematisch verwandte Begriffe: Making, Agent, Refuse, Hallucinate · 6 Treffer

Laden...

Videos werden geladen ...

Laden...

Beiträge werden geladen ...

Laden...

Videos werden geladen ...

Laden...

Beiträge werden geladen ...

Laden...

Videos werden geladen ...

Laden...

Beiträge werden geladen ...

Laden...

Videos werden geladen ...

Laden...

Beiträge werden geladen ...

Laden...

Videos werden geladen ...

Zum Aktualisieren ziehen
ZERO-DAY CVE-2026-94030 | A security vulnerability has been detected in SerenityOS up to 3d83e4509…
Advisory →
TTS Reader • tsecurity.de Voice
tsecurity.de Icon
tsecurity.de App
Offline-Lesen, Eilmeldungen & 0ms Ladezeit

Installiere tsecurity.de direkt auf deinen Home-Bildschirm für das ultimative Vollbild-Magazinerlebnis ohne Browser-Leisten.

Nächster Beitrag
Themen-Radar & Intelligence Matrix
Echtzeit-Taxonomie nach Angriffsvektoren & Plattformen

tsecurity.de Live Threat Radar

🔴 LIVE RADAR
MONITORING
AKTIV
CVE-DATENBANK
LIVE
🔍
Community Radar & Live Chat
Sentinel Bot online • Live-Stream
Dein Cluster: Security Explorer
Match:
lädt…
Verbindung zum Community-Stream wird aufgebaut...
Bearbeitungsmodus — Senden überschreibt deine Nachricht
Community-Puls — was gerade passiert
lädt…
Aktivitäten deiner Analysten
lädt…
Neues Thema oder Eilmeldung einreichen

Reiche interessante Links, Zero-Days oder Debatten ein. Die Community entscheidet per Upvote über die Veröffentlichung.

Heiß diskutierte Einreichungen
🔖 Gespeicherte Artikel
📂 Keine gespeicherten Artikel vorhanden.
Zurück Ziehen Vor
Links: vorheriger Artikel Rechts: nächster Artikel unten: schließen
News NIS-2 Frühwarnung Tier-1 Intel ⏱️ 3 Min vor 10 Min
Artikeldaten werden geladen...

Zurück: vorheriger Vor: nächster
↗ Original-Quelle
Social Reaktionen Deine Reaktion zählt
Einstufung & Relevanz-Poll 0 Stimmen
In sozialen Netzwerken teilen 1-Klick