Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences
Reviewed by Pithpith:VGPNUICVopen to challenge →
read the original abstract
Due to the cumbersome nature of human evaluation and limitations of code-based evaluation, Large Language Models (LLMs) are increasingly being used to assist humans in evaluating LLM outputs. Yet LLM-generated evaluators simply inherit all the problems of the LLMs they evaluate, requiring further human validation. We present a mixed-initiative approach to ``validate the validators'' -- aligning LLM-generated evaluation functions (be it prompts or code) with human requirements. Our interface, EvalGen, provides automated assistance to users in generating evaluation criteria and implementing assertions. While generating candidate implementations (Python functions, LLM grader prompts), EvalGen asks humans to grade a subset of LLM outputs; this feedback is used to select implementations that better align with user grades. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. In particular, we identify a phenomenon we dub \emph{criteria drift}: users need criteria to grade outputs, but grading outputs helps users define criteria. What is more, some criteria appears \emph{dependent} on the specific LLM outputs observed (rather than independent criteria that can be defined \emph{a priori}), raising serious questions for approaches that assume the independence of evaluation from observation of model outputs. We present our interface and implementation details, a comparison of our algorithm with a baseline approach, and implications for the design of future LLM evaluation assistants.
This paper has not been read by Pith yet.
Forward citations
Cited by 7 Pith papers
-
Results-Actionability Gap: Understanding How Practitioners Evaluate LLM Products in the Wild
Qualitative study of 19 practitioners reveals ten LLM product evaluation practices and introduces the results-actionability gap as a key barrier to turning findings into improvements.
-
Necessary but Not Sufficient: Temperature Control and Reproducibility in LLM-as-Judge Safety Evaluations
Even with temperature pinned at 0 and greedy decoding, 1-2 of 7 borderline safety items flip pass/fail verdicts across runs on real LLM providers, and some models deprecate temperature control entirely.
-
Toward Pre-Deployment Assurance for Enterprise AI Agents: Ontology-Grounded Simulation and Trust Certification
The authors introduce a three-part ontology-based verification system for AI agents that generates regulatory and adversarial test scenarios and issues machine-verifiable trust certificates, with pilot results indicat...
-
Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
Gemini 2.5 Flash with a Combined Budget debiasing strategy achieves 71.0% judge agreement at ~$0.001/evaluation, outperforming frontier models at 15x lower cost.
-
Judging the Judges: A Systematic Evaluation of Bias Mitigation Strategies in LLM-as-a-Judge Pipelines
Style bias dominates LLM-as-a-Judge systems far more than position bias, with debiasing strategies providing model-dependent gains and public tools released for replication.
-
Enhanced Self-Learning with Epistemologically-Informed LLM Dialogue
CausaDisco integrates Aristotle's Four Causes into LLM prompts to produce more engaging, exploratory, and multifaceted self-learning dialogues, as evidenced by controlled user studies.
-
Agentic-J: An AI Agent for Biological Microscopy Image Analysis
Agentic-J is a multi-agent AI assistant that converts natural language descriptions of biological image analysis tasks into executable, reproducible scripts for ImageJ/Fiji with specialised sub-agents for plugin manag...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.