Skip to content
by IQAC KIITKIIT Deemed to be University — Internal Quality Assurance Cell

Publication2024

Mitigating adversarial manipulation in LLMs: a prompt-based approach to counter Jailbreak attacks (Prompt-G)

The record

Type
Publication
Date
1 January 2024
Level
International — held abroad
Journal
Peerj Computer Science
Volume
10
Published
1 January 2024
DOI
10.7717/peerj-cs.2374
Scopus ID
85207915866
ISSN
23765992
Authors
7
Open access
Yes
Citations
21
International co-authorship
Yes
Co-author countries
IN, US

Classification

Every label below carries the reasoning it was assigned on, the confidence band it was assigned in, and what produced it. The rubric is published.

UN Sustainable Development Goals

  1. 16.3 Promote the rule of law at the national and international levels and ensure equal access to justice for all

    TargetPrimaryHigh confidenceJudged

    Adversarial attack mitigation using vector databases protects large language model integrity and security.

    High confidence
    The subject matter is a strong and direct inference from the record type, the funding or organising body, or the programme context, even where the title alone is thin. Example: an ICMR-funded project against SDG 3, on the strength of the agency.
    Judged
    Judged against the published classification rubric, under the rubric and prompt versions recorded with the run that produced it.
    Produced by
    claude-haiku-4-5-20251001, under kortex-publication-v1

  2. 17.6 Enhance North-South, South-South and triangular cooperation on science, technology and innovation, and knowledge sharing

    TargetSecondaryHigh confidenceRule-derivedrule sdg17-structural-v1

    Structural signal: co-authorship outside India.

    Rule-derived
    Emitted by a deterministic rule from structural signals rather than judged from the record’s text. The rule carries a version, so rule-derived and judged labels stay separable.
    Produced by
    sdg17-structural-v1, under kortex-publication-v1

Viksit Bharat @2047

  1. vb-4.4 Public safety, cybersecurity and national security

    Sub-themePrimaryHigh confidenceJudged

    LLM security framework development addresses cybersecurity challenges in AI systems.

    Produced by
    claude-haiku-4-5-20251001, under kortex-publication-v1

Each label above carries the instrument and the version that produced it. How each classification is reached.

Evidence

Kortex hosts no documents. Evidence is an outbound link to the authority that holds it.

No outbound evidence link is recorded for this activity. The record below states which source it came from and when it was last read.

Provenance

Which source system this record came from, how many source rows were resolved into it, and when each was last read.

Source rows resolved into this record
1

Each record carries the date Kortex last read it from its source system. That is a freshness date and not a verification: Kortex records what its sources say and when it read them, and it does not check those statements against any external authority. No claim on this site is described as verified, because none is.

Citing this record

This URL is stable. A record’s permalink is frozen when it is published and is not regenerated if its title is later corrected.

KIIT Deemed to be University. “Mitigating adversarial manipulation in LLMs: a prompt-based approach to counter Jailbreak attacks (Prompt-G)”. Kortex.

Explore the full record