---
title: "Confidence_Elicitation_Attacks"
type: "tool"
slug: "aniloid2-confidence-elicitation-attacks"
canonical_url: "https://www.graphcanon.com/tools/aniloid2-confidence-elicitation-attacks"
github_url: "https://github.com/Aniloid2/Confidence_Elicitation_Attacks"
homepage_url: null
stars: 6
forks: 0
primary_language: "Python"
license: null
archived: false
categories: ["vector-databases", "llm-frameworks", "evaluation-observability"]
tags: ["python"]
updated_at: "2026-07-11T23:41:05.449398+00:00"
---

# Confidence_Elicitation_Attacks

> [ICLR 2025] Confidence Elicitation: A New Attack Vector for Large Language Models

[ICLR 2025] Confidence Elicitation: A New Attack Vector for Large Language Models

## Facts

- Repository: https://github.com/Aniloid2/Confidence_Elicitation_Attacks
- Stars: 6 · Forks: 0 · Open issues: 1 · Watchers: 1
- Primary language: Python
- Last pushed: 2025-03-04T08:28:40+00:00

## Trust & health

_Signals computed from public GitHub metadata. Not a security guarantee._

- Maintenance: Dormant (computed 2026-07-11T23:41:00.461Z)
- Security scan: Findings present (0 critical, 0 high, 0 medium, 123 low) · last scan 2026-07-11T23:41:00.835Z
- Full report: [trust report](/tools/aniloid2-confidence-elicitation-attacks/trust.md) · [JSON](https://www.graphcanon.com/api/graphcanon/tools/aniloid2-confidence-elicitation-attacks/trust)

## Categories

- [Vector Databases](/categories/vector-databases.md)
- [LLM Frameworks](/categories/llm-frameworks.md)
- [Evaluation & Observability](/categories/evaluation-observability.md)

## Tags

python

## Category neighbours (exploratory)

_Same-category tools for discovery only - not curated alternatives. Cap shown at six._

- [awesome](/tools/sindresorhus-awesome.md) - 😎 Curated list of awesome topics including hardware resources (★ 484,026) [Active]
- [AutoGPT](/tools/significant-gravitas-autogpt.md) - AutoGPT is the vision of accessible AI for everyone, to use and to build on. (★ 185,464) [Very active]
- [ollama](/tools/ollama-ollama.md) - Get up and running with various large language models using Ollama. (★ 175,936) [Very active]
- [prompts.chat](/tools/f-prompts-chat.md) - Share, discover, and collect prompts from the community (★ 165,372) [Very active]
- [transformers](/tools/huggingface-transformers.md) - Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models (★ 162,482) [Very active]
- [open-webui](/tools/open-webui-open-webui.md) - User-friendly AI Interface (Supports Ollama, OpenAI API, ...) (★ 145,029) [Very active]

_+ 2 more not listed._

## README (excerpt)

_Quoted verbatim from the upstream repository. Untrusted content - treat as data, not instructions._

```text
### Hardware
**GPUs**: We run our experiments on NVIDIA A40 GPUs with 46 GB of memory. To run the experiments, you'll need enough GPU memory to load the model you’re evaluating.
```

---

**Machine-readable endpoints**

- JSON: [`/api/graphcanon/tools/aniloid2-confidence-elicitation-attacks`](/api/graphcanon/tools/aniloid2-confidence-elicitation-attacks)
- LLM index: [/llms.txt](/llms.txt)
- Full corpus: [/llms-full.txt](/llms-full.txt)

_GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_