In Loving memory of Simon Mark Hughes...

Introduction

The HHEM model is an open source model, created by Vectara, for detecting hallucinations in LLMs. It is particularly useful in the context of building retrieval-augmented-generation (RAG) applications where a set of facts is summarized by an LLM, but the model can also be used in other contexts.

If you are interested to learn more about RAG or experiment with Vectara, you can sign up for a free Vectara account. Vectara now implements an improved version of HHEM which is a calibrated score with longer sequence length, and non-English language support. See more details here Now let's dive into the details of the model.

Cross-Encoder for Hallucination Detection

This model was trained using SentenceTransformers Cross-Encoder class. The model outputs a probabilitity from 0 to 1, 0 being a hallucination and 1 being factually consistent. The predictions can be thresholded at 0.5 to predict whether a document is consistent with its source.

Training Data

This model is based on microsoft/deberta-v3-base and is trained initially on NLI data to determine textual entailment, before being further fine tuned on summarization datasets with samples annotated for factual consistency including FEVER, Vitamin C and PAWS.

Performance

TRUE Dataset (Minus Vitamin C, FEVER and PAWS) - 0.872 AUC Score
SummaC Benchmark (Test Split) - 0.764 Balanced Accuracy, 0.831 AUC Score
AnyScale Ranking Test for Hallucinations - 86.6 % Accuracy

LLM Hallucination Leaderboard

If you want to stay up to date with results of the latest tests using this model to evaluate the top LLM models, we have a public leaderboard that is periodically updated, and results are also available on the GitHub repository.

Using HHEM

Using the Inference API Widget on the Right

To use the model with the widget, you need to pass both documents as a single string separated with [SEP]. For example:

A man walks into a bar and buys a drink [SEP] A bloke swigs alcohol at a pub
A person on a horse jumps over a broken down airplane. [SEP] A person is at a diner, ordering an omelette.
A person on a horse jumps over a broken down airplane. [SEP] A person is outdoors, on a horse.

etc. See examples below for expected probability scores.

Usage with Sentencer Transformers (Recommended)

Inference

The model can be used like this, on pairs of documents, passed as a list of list of strings (List[List[str]]]):

from sentence_transformers import CrossEncoder

model = CrossEncoder('vectara/hallucination_evaluation_model')
scores = model.predict([
    ["A man walks into a bar and buys a drink", "A bloke swigs alcohol at a pub"],
    ["A person on a horse jumps over a broken down airplane.", "A person is at a diner, ordering an omelette."],
    ["A person on a horse jumps over a broken down airplane.", "A person is outdoors, on a horse."],
    ["A boy is jumping on skateboard in the middle of a red bridge.", "The boy skates down the sidewalk on a blue bridge"],
    ["A man with blond-hair, and a brown shirt drinking out of a public water fountain.", "A blond drinking water in public."],
    ["A man with blond-hair, and a brown shirt drinking out of a public water fountain.", "A blond man wearing a brown shirt is reading a book."],
    ["Mark Wahlberg was a fan of Manny.", "Manny was a fan of Mark Wahlberg."],  
])

This returns a numpy array representing a factual consistency score. A score < 0.5 indicates a likely hallucination):

array([0.61051559, 0.00047493709, 0.99639291, 0.00021221573, 0.99599433, 0.0014127002, 0.002.8262993], dtype=float32)

Note that the model is designed to work with entire documents, so long as they fit into the 512 token context window (across both documents). Also note that the order of the documents is important, the first document is the source document, and the second document is validated against the first for factual consistency, e.g. as a summary of the first or a claim drawn from the source.

Training

from sentence_transformers.cross_encoder import CrossEncoder
from sentence_transformers.cross_encoder.evaluation import CEBinaryClassificationEvaluator
from sentence_transformers import InputExample

num_epochs = 5
model_save_path = "./model_dump"
model_name = 'cross-encoder/nli-deberta-v3-base' # base model, use 'vectara/hallucination_evaluation_model' if you want to further fine-tune ours

model = CrossEncoder(model_name, num_labels=1, automodel_args={'ignore_mismatched_sizes':True})

# Load some training examples as such, using a pandas dataframe with source and summary columns:
train_examples, test_examples = [], []
for i, row in df_train.iterrows():
    train_examples.append(InputExample(texts=[row['source'], row['summary']], label=int(row['label'])))

for i, row in df_test.iterrows():
    test_examples.append(InputExample(texts=[row['source'], row['summary']], label=int(row['label'])))
test_evaluator = CEBinaryClassificationEvaluator.from_input_examples(test_examples, name='test_eval')

# Then train the model as such as per the Cross Encoder API:
train_dataloader = DataLoader(train_examples, shuffle=True, batch_size=train_batch_size)
warmup_steps = math.ceil(len(train_dataloader) * num_epochs * 0.1) #10% of train data for warm-up
model.fit(train_dataloader=train_dataloader,
          evaluator=test_evaluator,
          epochs=num_epochs,
          evaluation_steps=10_000,
          warmup_steps=warmup_steps,
          output_path=model_save_path,
          show_progress_bar=True)

Usage with Transformers AutoModel

You can use the model also directly with Transformers library (without the SentenceTransformers library):

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
import numpy as np

model = AutoModelForSequenceClassification.from_pretrained('vectara/hallucination_evaluation_model')
tokenizer = AutoTokenizer.from_pretrained('vectara/hallucination_evaluation_model')

pairs = [
    ["A man walks into a bar and buys a drink", "A bloke swigs alcohol at a pub"],
    ["A person on a horse jumps over a broken down airplane.", "A person is at a diner, ordering an omelette."],
    ["A person on a horse jumps over a broken down airplane.", "A person is outdoors, on a horse."],
    ["A boy is jumping on skateboard in the middle of a red bridge.", "The boy skates down the sidewalk on a blue bridge"],
    ["A man with blond-hair, and a brown shirt drinking out of a public water fountain.", "A blond drinking water in public."],
    ["A man with blond-hair, and a brown shirt drinking out of a public water fountain.", "A blond man wearing a brown shirt is reading a book."],
    ["Mark Wahlberg was a fan of Manny.", "Manny was a fan of Mark Wahlberg."], 
]

inputs = tokenizer.batch_encode_plus(pairs, return_tensors='pt', padding=True)

model.eval()
with torch.no_grad():
    outputs = model(**inputs)
    logits = outputs.logits.cpu().detach().numpy()
    # convert logits to probabilities
    scores = 1 / (1 + np.exp(-logits)).flatten()

This returns a numpy array representing a factual consistency score. A score < 0.5 indicates a likely hallucination):

array([0.61051559, 0.00047493709, 0.99639291, 0.00021221573, 0.99599433, 0.0014127002, 0.002.8262993], dtype=float32)

Contact Details

Feel free to contact us with any questions:

X/Twitter - https://twitter.com/vectara or http://twitter.com/ofermend
Discussion forums
Discord server

For more information about Vectara and how to use our RAG-as-a-service API platform, check out our documentation.

vectara
/

hallucination_evaluation_model