datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
DecodingTrust
DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
Overview
This repo contains the source code of DecodingTrust. This research endeavor is designed to help researchers better understand the capabilities, limitations, and potential risks associated with deploying these state-of-the-art Large Language Models (LLMs). See our paper for details.
DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
Boxin Wang, Weixin Chen, Hengzhi… See the full description on the dataset page: https://huggingface.co/datasets/AI-Secure/DecodingTrust.SecureCodePairs
Dataset Summary
Field
Value
Version
1.2.0
License
MIT
Total code examples
470
LLM security trajectories
30
Languages (15)
Python, Java, JavaScript, TypeScript, Go, PHP, C#, Kotlin, Swift, Rust, Ruby, C, C++, Scala, YAML (Kubernetes)
Frameworks
Flask, Django, FastAPI, Spring Boot, Express, NestJS, Next.js, Laravel, ASP.NET Core, Gin, Android, iOS, Actix, Rails, Qt, Play, gRPC, GraphQL, Kubernetes
New in v1.2.0
+260 records (deep Python/Java packs… See the full description on the dataset page: https://huggingface.co/datasets/ismailtasdelen/SecureCodePairs.llm-trustworthy-leaderboard-resultsSecurePy150ksecurehealthiot-disease-dataset
SecureHealthIoT Cleaned Dataset
Cleaned symptom-disease dataset generated by Kaggle kernel run:
aryansingh21fd/securehealthiot-disease-trainer-v1.
MMDecodingTrust-T2I
Overview
This repo contains the text-to-image dataset of MMDT (Multimodal DecodingTrust). This research endeavor is designed to help researchers and practitioners better understand the capabilities, limitations, and potential risks involved in deploying the state-of-the-art Multimodal foundation models (MMFMs). This dataset focuses on the following six primary perspectives of trustworthiness, including safety, hallucination, fairness, privacy, adversarial robustness, and… See the full description on the dataset page: https://huggingface.co/datasets/AI-Secure/MMDecodingTrust-T2I.secure_codeptf-id-bench
PTF-ID-Bench
Progressive Trust Framework — Intelligent Disobedience Benchmark.
A 290-scenario benchmark testing whether an AI coding agent appropriately refuses harmful requests, complies with legitimate ones, and escalates ambiguous cases to a human.
Source repo: https://github.com/bdas-sec/ptf-id-bench
Leaderboard: https://bdas-sec.github.io/ptf-id-bench/
License: MIT
Categories
Category
Count
ADVERSARIAL
75
BOUNDARY
40
CLEAR_DANGER
55
CLEAR_SAFE… See the full description on the dataset page: https://huggingface.co/datasets/bdas-secure/ptf-id-bench.han-secure-identity-federation-dataset-v1
Humanoid Secure Identity Federation Dataset
This dataset models federated identity validation
across decentralized humanoid agents.
It captures authentication states,
credential verification layers,
trust scoring,
and cross-domain identity propagation.
Objective
To enable secure identity interoperability
within distributed humanoid ecosystems.
Data Fields
node_id
identity_hash
credential_signature
federation_domain_id
trust_score
verification_status… See the full description on the dataset page: https://huggingface.co/datasets/achiepatricia/han-secure-identity-federation-dataset-v1.