Papers › RepEval: Effective Text Evaluation with LLM Representation

RepEval: Effective Text Evaluation with LLM Representation

30 Apr 2024arXiv:2404.19563archive 2025-07-28

Shuqian Sheng, Yi Xu, Tianhang Zhang, Zanwei Shen, Luoyi Fu, Jiaxin Ding, Lei Zhou, Xiaoying Gan, Xinbing Wang, Chenghu Zhou

The era of Large Language Models (LLMs) raises new demands for automatic evaluation metrics, which should be adaptable to various application scenarios while maintaining low cost and effectiveness. Traditional metrics for automatic text evaluation are often tailored to specific scenarios, while LLM-based evaluation metrics are costly, requiring fine-tuning or rely heavily on the generation capabilities of LLMs. Besides, previous LLM-based metrics ignore the fact that, within the space of LLM representations, there exist direction vectors that indicate the estimation of text quality. To this end, we introduce RepEval, a metric that leverages the projection of LLM representations for evaluation. Through simple prompt modifications, RepEval can easily transition to various tasks, requiring only minimal sample pairs for direction vector construction. Results on fourteen datasets across two evaluation tasks demonstrate the high effectiveness of our method, which exhibits a higher correlation with human judgments than previous methods, even in complex evaluation scenarios involving pair-wise selection under nuanced aspects. Our work underscores the richness of information regarding text quality embedded within LLM representations, offering insights for the development of new metrics.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

The archive's official repository (susisheng/repeval) was not harvested; the 27 samples below come from 4 repositories the archive does not mark official.

Syntology 27 code samples harvested from 4 repositories linked to this paper: 18 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 3 honoured, 0 violated, 6 with no contract checked; 9 where Syntology's instrument failed) · 9 unverified. Where Syntology's instrument failed: 9 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 27 samples from 4 repositories, 18 ran (9 with no instrument failure; 9 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 18 samples that ran have no recorded date.

susisheng/repeval officialmentioned in paperpytorchSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

27 samples harvested: 18 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 3 honoured, 0 violated, 6 with no contract checked; 9 where Syntology's instrument failed) · 9 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
9ran · our draft was wrong
6ran
9unverified

Licence: 6 of the 27 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 4 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

neulab/BARTScore found in paper text by Syntology 1 sample, 0 ran · 1 unverified
BARTScorer bart_score.py unverified Apache-2.0 (permissive) · 3186cad3f4fa28a4 · report
shikib/usr found in paper text by Syntology 12 samples, 10 ran (5 instrument-failed) · 2 unverified
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
transformers/modeling_openai.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0f786c407fb1ee4c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
angle_defn transformers/modeling_ctrl.py ran Apache-2.0 (permissive) · 671a619bed5669a3 · report fixture: signature pos:Tensor/2/float, i:Tensor/1/int, d_model_size:int/0/int · returned Tensor · level recorded 2026-09-24
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
transformers/modeling_gpt2.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 8d23fbe2b99b840b · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
transformers/modeling_bert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 211753ba29188d4e · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input
transformers/modeling_distilbert.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 69c6d14de8190cfb · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-24
gelu_new in 55 papers
census: 3 implementations from 7 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
transformers/modeling_bert.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 77601724cad03f95 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
load_tf_weights_in_bert in 135 papers transformers/modeling_bert.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · 0c145728a35789fa · report fixture: signature model:callable/0/?, config:Config/0/?, tf_checkpoint_path:str/0/str · returned BertModel · level recorded 2026-09-22
positional_encoding in 63 papers
census: 5 implementations from 6 papers, 4 distinct outputs on one shared input
transformers/modeling_ctrl.py ran Apache-2.0 (permissive) · 59ac77747a12b1cb · report fixture: signature position:int/0/int, d_model_size:int/0/int, dtype:dtype/0/? · returned Tensor · level recorded 2026-09-24
scaled_dot_product_attention in 43 papers transformers/modeling_ctrl.py ran Apache-2.0 (permissive) · 57ebc1447d921c46 · report fixture: signature q:Tensor/4/float, k:Tensor/4/float, v:Tensor/4/float, mask:Tensor/4/float, attention_mask:Tensor/4/float, head_mask:Tensor/4/float · returned tuple · level recorded 2026-09-24
tile in 109 papers
census: 5 implementations from 7 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
transformers/modeling_beam_search.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · a878dc3112f9ce9e · report fixture: signature x:Tensor/2/float, count:int/0/int, dim:int/0/int · returned Tensor · level recorded 2026-09-24
load_tf_weights_in_gpt2 in 68 papers transformers/modeling_gpt2.py unverified Apache-2.0 (permissive) · 654e91efc679fe84 · report
load_tf_weights_in_openai_gpt in 13 papers transformers/modeling_openai.py unverified Apache-2.0 (permissive) · 42ee168c3ae12e60 · report
jinlanfu/GPTScore found in paper text by Syntology 3 samples, 0 ran · 3 unverified
GPT3Model score_d2t.py unverified no licence file found · pointer only · 388a367eded0b858 · report
Scorer score_d2t.py unverified no licence file found · pointer only · 70093442b3ee191f · report
gpt3score score_d2t.py unverified no licence file found · pointer only · 6b02658358cb6a62 · report
maszhongming/UniEval found in paper text by Syntology 11 samples, 8 ran (4 instrument-failed) · 3 unverified
add_question utils.py ran MIT (permissive) · 1a5cb15d63f16098 · report fixture: signature dimension:str/0/str, output:list/1/str, src:list/1/str, ref:list/1/str, context:list/1/str, task:str/0/str · returned list · level recorded 2026-09-24
calculate_correlation in 5 papers reproduce/correlation.py ran MIT (permissive) · 31b107371370f1bf · report fixture: signature pred_score:list/1/float, human_score:list/1/float, dim:str/0/str, result:dict/0/? · returned dict · level recorded 2026-09-24
convert_to_json in 6 papers utils.py ran MIT (permissive) · 40a6d254f3b50b99 · report fixture: signature output_list:list/1/str, src_list:list/1/str, ref_list:list/1/str, context_list:list/1/str, scores:list/1/dict, doc_id:list/1/int, system_id:list/1/int · returned list · level recorded 2026-09-24
get_unique_value reproduce/correlation.py ran · honoured contract MIT (permissive) · 729c770d517b1bc6 · report fixture: signature data:list/1/dict, key:str/0/str · returned list · level recorded 2026-09-24
load_data in 700 papers pseudo_data_summ.py ran · our draft was wrong MIT (permissive) · cdc21dc661030083 · report fixture: signature data_path:str/0/str · returned list · level recorded 2026-09-24
load_json in 344 papers reproduce/data_utils.py ran · our draft was wrong MIT (permissive) · 79ec7dfe27ceef16 · report fixture: signature data_path:str/0/str · returned dict · level recorded 2026-09-24
load_output billboard/evaluate.py ran · our draft was wrong MIT (permissive) · 154c40d3e5f52366 · report fixture: signature output_path:str/0/str · returned list · level recorded 2026-09-24
load_src billboard/evaluate.py ran · our draft was wrong MIT (permissive) · 7f6c2b0cc66bbc7c · report fixture: signature src_path:str/0/str · returned list · level recorded 2026-09-24
disfluency_transformation pseudo_data_summ.py unverified MIT (permissive) · 666cfad4555e9faa · report
incoherence_transformation pseudo_data_summ.py unverified MIT (permissive) · bb6e70aa3a9fcf4e · report
load_ref billboard/evaluate.py unverified MIT (permissive) · dbd9587f7a986fc3 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for RepEval (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for RepEval (graph method: RepEval). 3 bricks; individually on a fixture: 0 ran · 3 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. Scorer entry · level 0 · unverified · 70093442b3ee191f
  2. gpt3score model · level 0 · unverified · 6b02658358cb6a62
  3. GPT3Model helper · level 0 · unverified · 388a367eded0b858

harvested Assembly for RepEval (graph method: RepEval). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. BARTScorer entry · level 0 · unverified · 3186cad3f4fa28a4

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2404.19563") returns these assemblies with each brick's role and wiring (how to connect).

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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