forked from microsoft/agent-governance-toolkit
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathCITATION.cff
More file actions
48 lines (46 loc) · 1.65 KB
/
CITATION.cff
File metadata and controls
48 lines (46 loc) · 1.65 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
# CITATION.cff - Citation File Format for academic citations
# https://citation-file-format.github.io/
# This file enables "Cite this repository" on GitHub
cff-version: 1.2.0
title: "CMVK — Verification Kernel"
message: "If you use this software in your research, please cite it as below."
type: software
authors:
- family-names: "Siddique"
given-names: "Imran"
orcid: "https://orcid.org/0000-0000-0000-0000" # Update with your ORCID
repository-code: "https://github.com/microsoft/agent-governance-toolkit"
url: "https://github.com/microsoft/agent-governance-toolkit"
license: MIT
version: "0.1.0"
date-released: "2026-01-01"
keywords:
- verification
- ai-safety
- hallucination-detection
- drift-detection
- adversarial
- multi-model
- code-generation
- llm
abstract: >-
CMVK is an adversarial multi-model verification framework that addresses
the "grading your own homework" fallacy in AI systems. By employing
strategic model diversity with separate Generator and Verifier roles,
CMVK reduces correlated blind spots and detects hallucinations that
self-verification would miss.
# Preferred citation for the paper (update when published)
preferred-citation:
type: conference-paper
authors:
- family-names: "Siddique"
given-names: "Imran"
title: "CMVK — Verification Kernel: Adversarial Multi-Model Code Generation with Blind Spot Reduction"
year: 2026
collection-title: "Advances in Neural Information Processing Systems"
collection-type: proceedings
conference:
name: "NeurIPS 2026"
# Update these fields when the paper is accepted/published:
# url: "https://proceedings.neurips.cc/paper/..."
# doi: "10.xxxx/..."