EvalAgent: Discovering Implicit Evaluation Criteria from the Web

• URL: http://arxiv.org/abs/2504.15219v1
• Date: Mon, 21 Apr 2025 16:43:50 GMT
• Title: EvalAgent: Discovering Implicit Evaluation Criteria from the Web
• Authors: Manya Wadhwa, Zayne Sprague, Chaitanya Malaviya, Philippe Laban, Junyi Jessy Li, Greg Durrett,
• Abstract summary: We introduce EvalAgent, a framework designed to automatically uncover nuanced and task-specific criteria.<n>EvalAgent mines expert-authored online guidance to propose diverse, long-tail evaluation criteria.<n>Our experiments demonstrate that the grounded criteria produced by EvalAgent are often implicit, yet specific.
• Score: 82.82096383262068
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Evaluation of language model outputs on structured writing tasks is typically conducted with a number of desirable criteria presented to human evaluators or large language models (LLMs). For instance, on a prompt like "Help me draft an academic talk on coffee intake vs research productivity", a model response may be evaluated for criteria like accuracy and coherence. However, high-quality responses should do more than just satisfy basic task requirements. An effective response to this query should include quintessential features of an academic talk, such as a compelling opening, clear research questions, and a takeaway. To help identify these implicit criteria, we introduce EvalAgent, a novel framework designed to automatically uncover nuanced and task-specific criteria. EvalAgent first mines expert-authored online guidance. It then uses this evidence to propose diverse, long-tail evaluation criteria that are grounded in reliable external sources. Our experiments demonstrate that the grounded criteria produced by EvalAgent are often implicit (not directly stated in the user's prompt), yet specific (high degree of lexical precision). Further, EvalAgent criteria are often not satisfied by initial responses but they are actionable, such that responses can be refined to satisfy them. Finally, we show that combining LLM-generated and EvalAgent criteria uncovers more human-valued criteria than using LLMs alone.

Related papers

• CritiQ: Mining Data Quality Criteria from Human Preferences [70.35346554179036]
  We introduce CritiQ, a novel data selection method that automatically mines criteria from human preferences for data quality.<n>CritiQ Flow employs a manager agent to evolve quality criteria and worker agents to make pairwise judgments.<n>We demonstrate the effectiveness of our method in the code, math, and logic domains.
  arXiv  Detail & Related papers  (2025-02-26T16:33:41Z)
• Contextualized Evaluations: Taking the Guesswork Out of Language Model Evaluations [85.81295563405433]
  Language model users often issue queries that lack specification, where the context under which a query was issued is not explicit. We present contextualized evaluations, a protocol that synthetically constructs context surrounding an under-specified query and provides it during evaluation. We find that the presence of context can 1) alter conclusions drawn from evaluation, even flipping win rates between model pairs, 2) nudge evaluators to make fewer judgments based on surface-level criteria, like style, and 3) provide new insights about model behavior across diverse contexts.
  arXiv  Detail & Related papers  (2024-11-11T18:58:38Z)
• TALEC: Teach Your LLM to Evaluate in Specific Domain with In-house Criteria by Criteria Division and Zero-shot Plus Few-shot [2.186726107112913]
  We propose a model-based evaluation method: TALEC. It allows users to flexibly set their own evaluation criteria, and uses in-context learning (ICL) to teach judge model these in-house criteria. TALEC demonstrates a strong capability to accurately reflect human preferences and achieves a correlation of over 80% with human judgments.
  arXiv  Detail & Related papers  (2024-06-25T10:02:42Z)
• Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences [11.23629471911503]
  EvalGen provides automated assistance to users in generating evaluation criteria and implementing assertions. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. We identify a phenomenon we dub emphcriteria drift: users need criteria to grade outputs, but grading outputs helps users define criteria.
  arXiv  Detail & Related papers  (2024-04-18T15:45:27Z)
• LLMCRIT: Teaching Large Language Models to Use Criteria [38.12026374220591]
  We propose a framework that enables large language models (LLMs) to use comprehensive criteria for a task in delivering natural language feedback on task execution. In particular, we present a model-in-the-loop framework that semi-automatically derives criteria from collected guidelines for different writing tasks and constructs in-context demonstrations for each criterion. The results reveal the fine-grained effects of incorporating criteria and demonstrations and provide valuable insights on how to teach LLMs to use criteria more effectively.
  arXiv  Detail & Related papers  (2024-03-02T02:25:55Z)
• HD-Eval: Aligning Large Language Model Evaluators Through Hierarchical Criteria Decomposition [92.17397504834825]
  HD-Eval is a framework that iteratively aligns large language models evaluators with human preference. HD-Eval inherits the essence from the evaluation mindset of human experts and enhances the alignment of LLM-based evaluators. Extensive experiments on three evaluation domains demonstrate the superiority of HD-Eval in further aligning state-of-the-art evaluators.
  arXiv  Detail & Related papers  (2024-02-24T08:01:32Z)
• PROXYQA: An Alternative Framework for Evaluating Long-Form Text Generation with Large Language Models [72.57329554067195]
  ProxyQA is an innovative framework dedicated to assessing longtext generation. It comprises in-depth human-curated meta-questions spanning various domains, each accompanied by specific proxy-questions with pre-annotated answers. It assesses the generated content's quality through the evaluator's accuracy in addressing the proxy-questions.
  arXiv  Detail & Related papers  (2024-01-26T18:12:25Z)
• Exploring the Reliability of Large Language Models as Customized Evaluators for Diverse NLP Tasks [65.69651759036535]
  We analyze whether large language models (LLMs) can serve as reliable alternatives to humans.<n>This paper explores both conventional tasks (e.g., story generation) and alignment tasks (e.g., math reasoning)<n>We find that LLM evaluators can generate unnecessary criteria or omit crucial criteria, resulting in a slight deviation from the experts.
  arXiv  Detail & Related papers  (2023-10-30T17:04:35Z)
• EvalLM: Interactive Evaluation of Large Language Model Prompts on User-Defined Criteria [43.944632774725484]
  We present EvalLM, an interactive system for iteratively refining prompts by evaluating multiple outputs on user-defined criteria. By describing criteria in natural language, users can employ the system's LLM-based evaluator to get an overview of where prompts excel or fail. A comparative study showed that EvalLM, when compared to manual evaluation, helped participants compose more diverse criteria, examine twice as many outputs, and reach satisfactory prompts with 59% fewer revisions.
  arXiv  Detail & Related papers  (2023-09-24T13:19:38Z)

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.