HallucinationDetector

HallucinationDetector solves the problem of models making up answers by highlighting the answers that don't exist in the context.

Suggest Edits

📘
Changes on their way
We're still actively working on this feature to make it better. This page describes its current, first implementation. We'll be updating it soon to make it smoother.

HallucinationDetector is designed for retrieval-augmented generative question answering systems, where the model generates answers based on your documents. To avoid hallucinations, HallucinationDetector uses a model to indicate the answers that are grounded in the documents and the ones that are not.

If you add HallucinationDetector to your generative pipeline, you can then enable it on the Search page. When enabled. HallucinationDetector highlights each sentence in the answer, indicating if it's grounded in the documents. Green highlight means the sentence is based on the documents. Orange highlight means it's partially based on the documents. Red highlight means it's a hallucination.

An example of an answer with hallucinations underlined

Basic Information

Pipeline type: Used in query pipelines.
Position in a pipeline: After PromptNode.
Input: Documents
Output: Answers
Available Classes: TransformersHallucinationDetector

Usage Example

It's best to leave the node with its default parameters:

...
components:
	- name: HallucinationDetector
    type: TransformersHallucinationDetector
...
pipelines:
  - name: query
    nodes:
      - name: Retriever
        inputs: [Query]
      - name: PromptNode
        inputs: [Retriever]
      - name: HallucinationDetector
        inputs: [PromptNode]
...

Arguments

Argument	Type	Possible Values	Description
`model_name_or_path`	String	It uses a model we trained for detecting hallucinations.	The model used for detecting whether an answer is grounded in the documents. Mandatory.
`model_version`	String		The version of the model. Optional.
`language`	String	`cs` - Czech `da` - Danish `de` - German `el` - Greek `en` - English `es` - Spanish `et` - Estonian `fi` - Finnish `fr` - French `it` - Italian `ml` - Malayam `nl` - Dutch `no` - Norwegian `pl` - Polish `pt` - Portugese `ru` - Russian `sl` - Slovene `sv` - Swedish `tr` - Turkish Default: `en`	The language of the documents using the nltk format. Mandatory.
`use_gpu`	Boolean	True/False Default: `True`	Uses GPU if available. If not, it falls back on the CPU. Mandatory.
`batch_size`	Integer	Default: `16`	The batch size for inference. Mandatory.
`use_auth_token`	String, Boolean		An API token if you're using a private model. Optional.
`devices`	String, torch.device		A list of devices to use for inference. Optional.
`class_thresholds`	String, float	Default: `"full_support": 1.5` `"partial_support": -0.5` `"no_support": -0.2`	The thresholds for the classes. The classes are: - `full_support` (the whole sentence is grounded in the documents) - `partial_support` (a part of the sentence is grounded in the documents) - `no_support` (the sentence is not in the documents) - `contradiction` (the sentence contradicts the documents). We trained the model to return: `2` for full_support `1` for partial_support `0` for no_support `-1` for contradiction. Thresholds are inclusive, which means that if `full_support` is `1.3`, the value of `1.3` is considered `full_support`. Changing the thresholds between the classes can help you calibrate what the model considers as grounded in the documents or not. Optional.
`default_class`	String	`contradiction` `full_support` `no_support` `partial_support` Default: `contradiction`	The default class if the value is below all thresholds. Mandatory.

Updated about 1 year ago

📘Changes on their way

Basic Information

Usage Example

Arguments

📘
Changes on their way