Papers › Are LLM-based Evaluators Confusing NLG Quality Criteria?

Are LLM-based Evaluators Confusing NLG Quality Criteria?

19 Feb 2024arXiv:2402.12055archive 2025-07-28

Xinyu Hu, Mingqi Gao, Sen Hu, Yang Zhang, Yicheng Chen, Teng Xu, Xiaojun Wan

Some prior work has shown that LLMs perform well in NLG evaluation for different tasks. However, we discover that LLMs seem to confuse different evaluation criteria, which reduces their reliability. For further verification, we first consider avoiding issues of inconsistent conceptualization and vague expression in existing NLG quality criteria themselves. So we summarize a clear hierarchical classification system for 11 common aspects with corresponding different criteria from previous studies involved. Inspired by behavioral testing, we elaborately design 18 types of aspect-targeted perturbation attacks for fine-grained analysis of the evaluation behaviors of different LLMs. We also conduct human annotations beyond the guidance of the classification system to validate the impact of the perturbations. Our experimental results reveal confusion issues inherent in LLMs, as well as other noteworthy phenomena, and necessitate further research and improvements for LLM-based evaluation.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2402.12055")

Code

Syntology Ran 3 of 8 code samples harvested from 1 repository linked to this paper; 5 have no recorded run. Of those that ran: 3 ran with no contract checked.

By repository: official repository: 8 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

herrxy/llm-evaluator-reliability officialmentioned in papermentioned on GitHubApache-2.0 report
pku-onelab/llm-evaluator-reliability officialmentioned in papermentioned on GitHubApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

8 samples harvested; 3 ran; 0 honoured the contract we drafted; 5 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

3ran
5unverified

Licence: 0 of the 8 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from herrxy/llm-evaluator-reliability. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

contains_non_alpha herrxy/llm-evaluator-reliability/Perturbations/grammar_order.py official repository ran Apache-2.0 (permissive) · 1e3e86a7baafca68 · report
introduce_swap herrxy/llm-evaluator-reliability/Perturbations/grammar_order.py official repository ran fingerprinted Apache-2.0 (permissive) · 26336d4815eced14 · report
introduce_typo herrxy/llm-evaluator-reliability/Perturbations/grammar_typo.py official repository ran fingerprinted Apache-2.0 (permissive) · 97382afbeed4ecb1 · report
attack_coherence_sentence_order herrxy/llm-evaluator-reliability/Perturbations/coherence_order.py official repository unverified Apache-2.0 (permissive) · a344135785b6fb9b · report
attack_grammar_typos herrxy/llm-evaluator-reliability/Perturbations/grammar_typo.py official repository unverified Apache-2.0 (permissive) · acb43e2bc12c97cc · report
attack_grammar_word_order herrxy/llm-evaluator-reliability/Perturbations/grammar_order.py official repository unverified Apache-2.0 (permissive) · 05e821e68b4582a4 · report
attack_informative_delete_last herrxy/llm-evaluator-reliability/Perturbations/inform_delete.py official repository unverified Apache-2.0 (permissive) · 1d690ef0ad6fdfc4 · report
is_the_same herrxy/llm-evaluator-reliability/Perturbations/coherence_order.py official repository unverified Apache-2.0 (permissive) · 90afb16f86fbb016 · report

Tasks

nlg evaluation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections