Benchmarking Llama2, Mistral, Gemma and GPT for Factuality, Toxicity, Bias and Propensity for Hallucinations

• URL: http://arxiv.org/abs/2404.09785v1
• Date: Mon, 15 Apr 2024 13:40:08 GMT
• Title: Benchmarking Llama2, Mistral, Gemma and GPT for Factuality, Toxicity, Bias and Propensity for Hallucinations
• Authors: David Nadeau, Mike Kroutikov, Karen McNeil, Simon Baribeau,
• Abstract summary: This paper introduces fourteen novel datasets for the evaluation of Large Language Models' safety in the context of enterprise tasks. A method was devised to evaluate a model's safety, as determined by its ability to follow instructions and output factual, unbiased, grounded, and appropriate content.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This paper introduces fourteen novel datasets for the evaluation of Large Language Models' safety in the context of enterprise tasks. A method was devised to evaluate a model's safety, as determined by its ability to follow instructions and output factual, unbiased, grounded, and appropriate content. In this research, we used OpenAI GPT as point of comparison since it excels at all levels of safety. On the open-source side, for smaller models, Meta Llama2 performs well at factuality and toxicity but has the highest propensity for hallucination. Mistral hallucinates the least but cannot handle toxicity well. It performs well in a dataset mixing several tasks and safety vectors in a narrow vertical domain. Gemma, the newly introduced open-source model based on Google Gemini, is generally balanced but trailing behind. When engaging in back-and-forth conversation (multi-turn prompts), we find that the safety of open-source models degrades significantly. Aside from OpenAI's GPT, Mistral is the only model that still performed well in multi-turn tests.

Related papers

• ShieldGemma: Generative AI Content Moderation Based on Gemma [49.91147965876678]
  ShieldGemma is a suite of safety content moderation models built upon Gemma2. Models provide robust, state-of-the-art predictions of safety risks across key harm types.
  arXiv  Detail & Related papers  (2024-07-31T17:48:14Z)
• What Makes and Breaks Safety Fine-tuning? A Mechanistic Study [64.9691741899956]
  Safety fine-tuning helps align Large Language Models (LLMs) with human preferences for their safe deployment. We design a synthetic data generation framework that captures salient aspects of an unsafe input. Using this, we investigate three well-known safety fine-tuning methods.
  arXiv  Detail & Related papers  (2024-07-14T16:12:57Z)
• MetaCheckGPT -- A Multi-task Hallucination Detector Using LLM Uncertainty and Meta-models [8.322071110929338]
  This paper describes our winning solution ranked 1st and 2nd in the 2 sub-tasks of model agnostic and model aware tracks respectively. We propose a meta-regressor framework of LLMs for model evaluation and integration that achieves the highest scores on the leaderboard.
  arXiv  Detail & Related papers  (2024-04-10T11:56:01Z)
• Gemma: Open Models Based on Gemini Research and Technology [128.57714343844074]
  This work introduces Gemma, a family of lightweight, state-of-the art open models built from the research and technology used to create Gemini models. Gemma models demonstrate strong performance across academic benchmarks for language understanding, reasoning, and safety.
  arXiv  Detail & Related papers  (2024-03-13T06:59:16Z)
• Keeping LLMs Aligned After Fine-tuning: The Crucial Role of Prompt Templates [59.0123809721502]
  This paper proposes the "Pure Tuning, Safe Testing" (PTST) principle -- fine-tune models without a safety prompt, but include it at test time. Fine-tuning experiments on GSM8K, ChatDoctor, and OpenOrca show that PTST significantly reduces the rise of unsafe behaviors.
  arXiv  Detail & Related papers  (2024-02-28T18:23:49Z)
• Emulated Disalignment: Safety Alignment for Large Language Models May Backfire! [65.06450319194454]
  Large language models (LLMs) undergo safety alignment to ensure safe conversations with humans. This paper introduces a training-free attack method capable of reversing safety alignment. We name this method emulated disalignment (ED) because sampling from this contrastive distribution provably emulates the result of fine-tuning to minimize a safety reward.
  arXiv  Detail & Related papers  (2024-02-19T18:16:51Z)
• Language Model Unalignment: Parametric Red-Teaming to Expose Hidden Harms and Biases [32.2246459413988]
  Red-teaming aims to jailbreak a model's safety behavior to make it act as a helpful agent disregarding the harmfulness of the query. We present a new perspective on safety research i.e., red-teaming through Unalignment. Unalignment tunes the model parameters to break model guardrails that are not deeply rooted in the model's behavior.
  arXiv  Detail & Related papers  (2023-10-22T13:55:46Z)
• Safe MDP Planning by Learning Temporal Patterns of Undesirable Trajectories and Averting Negative Side Effects [27.41101006357176]
  In safe MDP planning, a cost function based on the current state and action is often used to specify safety aspects. operating based on an incomplete model can often produce unintended negative side effects (NSEs)
  arXiv  Detail & Related papers  (2023-04-06T14:03:24Z)
• LaMDA: Language Models for Dialog Applications [75.75051929981933]
  LaMDA is a family of Transformer-based neural language models specialized for dialog. Fine-tuning with annotated data and enabling the model to consult external knowledge sources can lead to significant improvements.
  arXiv  Detail & Related papers  (2022-01-20T15:44:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.