KIEval: A Knowledge-grounded Interactive Evaluation Framework for Large Language Models
- URL: http://arxiv.org/abs/2402.15043v2
- Date: Mon, 3 Jun 2024 06:02:39 GMT
- Title: KIEval: A Knowledge-grounded Interactive Evaluation Framework for Large Language Models
- Authors: Zhuohao Yu, Chang Gao, Wenjin Yao, Yidong Wang, Wei Ye, Jindong Wang, Xing Xie, Yue Zhang, Shikun Zhang,
- Abstract summary: KIEval is a Knowledge-grounded Interactive Evaluation framework for large language models.
It incorporates an LLM-powered "interactor" role for the first time to accomplish a dynamic contamination-resilient evaluation.
Extensive experiments on seven leading LLMs across five datasets validate KIEval's effectiveness and generalization.
- Score: 53.84677081899392
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Automatic evaluation methods for large language models (LLMs) are hindered by data contamination, leading to inflated assessments of their effectiveness. Existing strategies, which aim to detect contaminated texts, focus on quantifying contamination status instead of accurately gauging model performance. In this paper, we introduce KIEval, a Knowledge-grounded Interactive Evaluation framework, which incorporates an LLM-powered "interactor" role for the first time to accomplish a dynamic contamination-resilient evaluation. Starting with a question in a conventional LLM benchmark involving domain-specific knowledge, KIEval utilizes dynamically generated, multi-round, and knowledge-focused dialogues to determine whether a model's response is merely a recall of benchmark answers or demonstrates a deep comprehension to apply knowledge in more complex conversations. Extensive experiments on seven leading LLMs across five datasets validate KIEval's effectiveness and generalization. We also reveal that data contamination brings no contribution or even negative effect to models' real-world applicability and understanding, and existing contamination detection methods for LLMs can only identify contamination in pre-training but not during supervised fine-tuning.
Related papers
- Benchmark Data Contamination of Large Language Models: A Survey [5.806534973464769]
This paper reviews the complex challenge of Benchmark Data Contamination (BDC) in Large Language Models (LLMs) evaluation.
It explores alternative assessment methods to mitigate the risks associated with traditional benchmarks.
arXiv Detail & Related papers (2024-06-06T16:41:39Z) - Unveiling the Misuse Potential of Base Large Language Models via In-Context Learning [61.2224355547598]
Open-sourcing of large language models (LLMs) accelerates application development, innovation, and scientific progress.
Our investigation exposes a critical oversight in this belief.
By deploying carefully designed demonstrations, our research demonstrates that base LLMs could effectively interpret and execute malicious instructions.
arXiv Detail & Related papers (2024-04-16T13:22:54Z) - FreeEval: A Modular Framework for Trustworthy and Efficient Evaluation of Large Language Models [36.273451767886726]
FreeEval is a modular and scalable framework crafted to enable trustworthy and efficient automatic evaluations of large language models.
FreeEval's unified abstractions simplify the integration and improve the transparency of diverse evaluation methodologies.
The framework integrates meta-evaluation techniques like human evaluation and data contamination detection, which, along with dynamic evaluation modules, enhance the fairness of the evaluation outcomes.
arXiv Detail & Related papers (2024-04-09T04:17:51Z) - Evaluating Generative Language Models in Information Extraction as Subjective Question Correction [49.729908337372436]
We propose a new evaluation method, SQC-Score.
Inspired by the principles in subjective question correction, we propose a new evaluation method, SQC-Score.
Results on three information extraction tasks show that SQC-Score is more preferred by human annotators than the baseline metrics.
arXiv Detail & Related papers (2024-04-04T15:36:53Z) - Factual Consistency Evaluation of Summarisation in the Era of Large
Language Models [38.8292168447796]
Existing factual consistency metrics are constrained by their performance, efficiency, and explainability.
Recent advances in Large language models (LLMs) have demonstrated remarkable potential in text evaluation.
arXiv Detail & Related papers (2024-02-21T12:35:19Z) - ReEval: Automatic Hallucination Evaluation for Retrieval-Augmented Large Language Models via Transferable Adversarial Attacks [91.55895047448249]
This paper presents ReEval, an LLM-based framework using prompt chaining to perturb the original evidence for generating new test cases.
We implement ReEval using ChatGPT and evaluate the resulting variants of two popular open-domain QA datasets.
Our generated data is human-readable and useful to trigger hallucination in large language models.
arXiv Detail & Related papers (2023-10-19T06:37:32Z) - Bring Your Own Data! Self-Supervised Evaluation for Large Language
Models [52.15056231665816]
We propose a framework for self-supervised evaluation of Large Language Models (LLMs)
We demonstrate self-supervised evaluation strategies for measuring closed-book knowledge, toxicity, and long-range context dependence.
We find strong correlations between self-supervised and human-supervised evaluations.
arXiv Detail & Related papers (2023-06-23T17:59:09Z) - Evaluation Gaps in Machine Learning Practice [13.963766987258161]
In practice, evaluations of machine learning models frequently focus on a narrow range of decontextualized predictive behaviours.
We examine the evaluation gaps between the idealized breadth of evaluation concerns and the observed narrow focus of actual evaluations.
By studying these properties, we demonstrate the machine learning discipline's implicit assumption of a range of commitments which have normative impacts.
arXiv Detail & Related papers (2022-05-11T04:00:44Z) - Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy
Evaluation Approach [84.02388020258141]
We propose a new framework named ENIGMA for estimating human evaluation scores based on off-policy evaluation in reinforcement learning.
ENIGMA only requires a handful of pre-collected experience data, and therefore does not involve human interaction with the target policy during the evaluation.
Our experiments show that ENIGMA significantly outperforms existing methods in terms of correlation with human evaluation scores.
arXiv Detail & Related papers (2021-02-20T03:29:20Z)
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.