Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing
- URL: http://arxiv.org/abs/2406.14230v2
- Date: Fri, 12 Jul 2024 03:47:21 GMT
- Title: Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing
- Authors: Han Jiang, Xiaoyuan Yi, Zhihua Wei, Shu Wang, Xing Xie,
- Abstract summary: Large Language Models (LLMs) have achieved significant breakthroughs, but their generated unethical content poses potential risks.
Measuring value alignment of LLMs becomes crucial for their regulation and responsible deployment.
We propose GETA, a novel generative evolving testing approach that dynamically probes the underlying moral baselines of LLMs.
- Score: 39.93490432227601
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Warning: this paper contains model outputs exhibiting unethical information. Large Language Models (LLMs) have achieved significant breakthroughs, but their generated unethical content poses potential risks. Measuring value alignment of LLMs becomes crucial for their regulation and responsible deployment. Numerous datasets have been constructed to assess social bias, toxicity, and ethics in LLMs, but they suffer from evaluation chronoeffect, that is, as models rapidly evolve, existing data becomes leaked or undemanding, overestimating ever-developing LLMs. To tackle this problem, we propose GETA, a novel generative evolving testing approach that dynamically probes the underlying moral baselines of LLMs. Distinct from previous adaptive testing methods that rely on static datasets with limited difficulty, GETA incorporates an iteratively-updated item generator which infers each LLM's moral boundaries and generates difficulty-tailored testing items, accurately reflecting the true alignment extent. This process theoretically learns a joint distribution of item and model response, with item difficulty and value conformity as latent variables, where the generator co-evolves with the LLM, addressing chronoeffect. We evaluate various popular LLMs with diverse capabilities and demonstrate that GETA can create difficulty-matching testing items and more accurately assess LLMs' values, better consistent with their performance on unseen OOD and i.i.d. items, laying the groundwork for future evaluation paradigms.
Related papers
- Estimating Item Difficulty Using Large Language Models and Tree-Based Machine Learning Algorithms [0.0]
Estimating item difficulty through field-testing is often resource-intensive and time-consuming.
The present research examines the feasibility of using an Large Language Models (LLMs) to predict item difficulty for K-5 mathematics and reading assessment items.
arXiv Detail & Related papers (2025-04-09T00:04:07Z) - BEATS: Bias Evaluation and Assessment Test Suite for Large Language Models [0.0]
We introduce BEATS, a novel framework for evaluating Bias, Ethics, Fairness, and Factuality in Large Language Models (LLMs)
We present a bias benchmark for LLMs that measure performance across 29 distinct metrics.
These metrics span a broad range of characteristics, including demographic, cognitive, and social biases, as well as measures of ethical reasoning, group fairness, and factuality related misinformation risk.
arXiv Detail & Related papers (2025-03-31T16:56:52Z) - Training an LLM-as-a-Judge Model: Pipeline, Insights, and Practical Lessons [9.954960702259918]
This paper introduces Themis, a fine-tuned large language model (LLMs) judge that delivers context-aware evaluations.
We provide a comprehensive overview of the development pipeline for Themis, highlighting its scenario-dependent evaluation prompts.
We introduce two human-labeled benchmarks for meta-evaluation, demonstrating that Themis can achieve high alignment with human preferences in an economical manner.
arXiv Detail & Related papers (2025-02-05T08:35:55Z) - The Vulnerability of Language Model Benchmarks: Do They Accurately Reflect True LLM Performance? [1.3810901729134184]
Large Language Models (LLMs) excel at standardized tests while failing to demonstrate genuine language understanding and adaptability.
Our systematic analysis of NLP evaluation frameworks reveals pervasive vulnerabilities across the evaluation spectrum.
We lay the groundwork for new evaluation methods that resist manipulation, minimize data contamination, and assess domain-specific tasks.
arXiv Detail & Related papers (2024-12-02T20:49:21Z) - Beyond Binary: Towards Fine-Grained LLM-Generated Text Detection via Role Recognition and Involvement Measurement [51.601916604301685]
Large language models (LLMs) generate content that can undermine trust in online discourse.
Current methods often focus on binary classification, failing to address the complexities of real-world scenarios like human-AI collaboration.
To move beyond binary classification and address these challenges, we propose a new paradigm for detecting LLM-generated content.
arXiv Detail & Related papers (2024-10-18T08:14:10Z) - DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph [70.79413606968814]
We introduce Dynamic Evaluation of LLMs via Adaptive Reasoning Graph Evolvement (DARG) to dynamically extend current benchmarks with controlled complexity and diversity.
Specifically, we first extract the reasoning graphs of data points in current benchmarks and then perturb the reasoning graphs to generate novel testing data.
Such newly generated test samples can have different levels of complexity while maintaining linguistic diversity similar to the original benchmarks.
arXiv Detail & Related papers (2024-06-25T04:27:53Z) - MoralBench: Moral Evaluation of LLMs [34.43699121838648]
This paper introduces a novel benchmark designed to measure and compare the moral reasoning capabilities of large language models (LLMs)
We present the first comprehensive dataset specifically curated to probe the moral dimensions of LLM outputs.
Our methodology involves a multi-faceted approach, combining quantitative analysis with qualitative insights from ethics scholars to ensure a thorough evaluation of model performance.
arXiv Detail & Related papers (2024-06-06T18:15:01Z) - Evaluating the Factuality of Large Language Models using Large-Scale Knowledge Graphs [30.179703001666173]
Factuality issue is a critical concern for Large Language Models (LLMs)
We propose GraphEval to evaluate an LLM's performance using a substantially large test dataset.
Test dataset is retrieved from a large knowledge graph with more than 10 million facts without expensive human efforts.
arXiv Detail & Related papers (2024-04-01T06:01:17Z) - CLOMO: Counterfactual Logical Modification with Large Language Models [109.60793869938534]
We introduce a novel task, Counterfactual Logical Modification (CLOMO), and a high-quality human-annotated benchmark.
In this task, LLMs must adeptly alter a given argumentative text to uphold a predetermined logical relationship.
We propose an innovative evaluation metric, the Self-Evaluation Score (SES), to directly evaluate the natural language output of LLMs.
arXiv Detail & Related papers (2023-11-29T08:29:54Z) - Don't Make Your LLM an Evaluation Benchmark Cheater [142.24553056600627]
Large language models(LLMs) have greatly advanced the frontiers of artificial intelligence, attaining remarkable improvement in model capacity.
To assess the model performance, a typical approach is to construct evaluation benchmarks for measuring the ability level of LLMs.
We discuss the potential risk and impact of inappropriately using evaluation benchmarks and misleadingly interpreting the evaluation results.
arXiv Detail & Related papers (2023-11-03T14:59:54Z) - 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) - Denevil: Towards Deciphering and Navigating the Ethical Values of Large
Language Models via Instruction Learning [36.66806788879868]
Large Language Models (LLMs) have made unprecedented breakthroughs, yet their integration into everyday life might raise societal risks due to generated unethical content.
This work delves into ethical values utilizing Moral Foundation Theory.
arXiv Detail & Related papers (2023-10-17T07:42:40Z) - DyVal: Dynamic Evaluation of Large Language Models for Reasoning Tasks [112.66827096358857]
We introduce DyVal, a protocol for dynamic evaluation of large language models (LLMs)
Based on our framework, we build graph-informed DyVal by leveraging the structural advantage of directed acyclic graphs.
We evaluate various LLMs ranging from Flan-T5-large to GPT-3.5-Turbo and GPT-4.
arXiv Detail & Related papers (2023-09-29T12:04:14Z) - Are Large Language Models Really Robust to Word-Level Perturbations? [68.60618778027694]
We propose a novel rational evaluation approach that leverages pre-trained reward models as diagnostic tools.
Longer conversations manifest the comprehensive grasp of language models in terms of their proficiency in understanding questions.
Our results demonstrate that LLMs frequently exhibit vulnerability to word-level perturbations that are commonplace in daily language usage.
arXiv Detail & Related papers (2023-09-20T09:23:46Z) - LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond [135.8013388183257]
We propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits.
Most LLMs struggle on SummEdits, with performance close to random chance.
The best-performing model, GPT-4, is still 8% below estimated human performance.
arXiv Detail & Related papers (2023-05-23T21:50:06Z)
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.