Papers › Agent-as-Judge for Factual Summarization of Long Narratives

Agent-as-Judge for Factual Summarization of Long Narratives

17 Jan 2025arXiv:2501.09993archive 2025-07-28

Yeonseok Jeong, Minsoo Kim, Seung-won Hwang, Byung-Hak Kim

Large Language Models (LLMs) have demonstrated near-human performance in summarization tasks based on traditional metrics such as ROUGE and BERTScore. However, these metrics do not adequately capture critical aspects of summarization quality, such as factual accuracy, particularly for long narratives (>100K tokens). Recent advances, such as LLM-as-a-Judge, address the limitations of metrics based on lexical similarity but still exhibit factual inconsistencies, especially in understanding character relationships and states. In this work, we introduce NarrativeFactScore, a novel "Agent-as-a-Judge" framework for evaluating and refining summaries. By leveraging a Character Knowledge Graph (CKG) extracted from input and generated summaries, NarrativeFactScore assesses the factual consistency and provides actionable guidance for refinement, such as identifying missing or erroneous facts. We demonstrate the effectiveness of NarrativeFactScore through a detailed workflow illustration and extensive validation on widely adopted benchmarks, achieving superior performance compared to competitive methods. Our results highlight the potential of agent-driven evaluation systems to improve the factual reliability of LLM-generated summaries.

PaperPDFCodeCode Syntology ran

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

Syntology 13 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 13 samples from 1 repository, 10 ran (10 with no instrument failure; 0 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.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

yeonseokjeong/narrativefactscore officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 13 samples, 10 ran (0 instrument-failed) · 3 unverified 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

13 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

10ran
3unverified

Licence: 0 of the 13 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 yeonseokjeong/narrativefactscore. “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.

yeonseokjeong/narrativefactscore official repository 13 samples, 10 ran (0 instrument-failed) · 3 unverified
break_down2scenes src/fact/utils.py ran fingerprinted Apache-2.0 (permissive) · 27358d1c0a6c53b6 · report fixture: signature text:str/0/str · returned list · level recorded 2026-09-24
build_fact_prompt src/fact/prompt.py ran Apache-2.0 (permissive) · 27c939504cb861df · report fixture: returned str · level recorded 2026-09-24
calculate_bert_score calculate_metrics.py ran Apache-2.0 (permissive) · a995b01f880808ea · report fixture: signature bert_scorer:MockBertScorer/0/?, reference_summary:str/0/str, generated_summary:str/0/str · returned tuple · level recorded 2026-09-24
calculate_rouge calculate_metrics.py ran Apache-2.0 (permissive) · 1ac357bf2eed797a · report fixture: signature reference_summary:str/0/str, generated_summary:str/0/str · returned tuple · level recorded 2026-09-24
delete_special src/fact/utils.py ran Apache-2.0 (permissive) · 2a13a9eae8815574 · report fixture: signature pre_text:str/0/str, character_list:list/1/str · returned str · level recorded 2026-09-24
generate_names_graph src/kg/knowledge_graph.py ran Apache-2.0 (permissive) · f50f1c74536e69c4 · report fixture: signature names:list/1/list · returned Graph · level recorded 2026-09-24
gpt_inference src/kg/generate_kg.py ran Apache-2.0 (permissive) · 00ab0223fa638a2f · report fixture: signature system_instruction:str/0/str, prompt:str/0/str, retries:int/0/int, delay:int/0/int · returned str · level recorded 2026-09-24
load_knowledge_graph src/kg/generate_kg.py ran Apache-2.0 (permissive) · c8df89ad35127fe7 · report fixture: signature project_gutenberg_id:int/0/int, save_path:PosixPath/0/? · returned dict · level recorded 2026-09-24
remove_number_prefix src/kg/knowledge_graph.py ran fingerprinted Apache-2.0 (permissive) · 8936fbeddd8acbb0 · report fixture: signature text:str/0/str · returned str · level recorded 2026-09-24
split_and_clean_script down_data.py ran fingerprinted Apache-2.0 (permissive) · 69d99f943fa8a011 · report fixture: signature script:str/0/str · returned list · level recorded 2026-09-24
calculate_bleu in 10 papers calculate_metrics.py unverified Apache-2.0 (permissive) · 50254ec8b6e5f2d1 · report
generate_knowledge_graph_for_scripts src/kg/generate_kg.py unverified Apache-2.0 (permissive) · d24be1d4cdb47e5f · report
parse_response_text src/kg/knowledge_graph.py unverified Apache-2.0 (permissive) · 4308a1d6439f3b09 · report

Tasks

Long-Form Narrative Summarization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Long-Form Narrative Summarization MENSA Hierarchically Merging and Agent Refinement BERTScore (F1) 60.22 #3 of 10 Archive leaderboard report
Long-Form Narrative Summarization MENSA Hierarchically Merging and Agent Refinement ROUGE-1 31.31 #3 of 10 Archive leaderboard report
Long-Form Narrative Summarization MENSA Hierarchically Merging and Agent Refinement ROUGE-2 8.81 #3 of 10 Archive leaderboard report
Long-Form Narrative Summarization MENSA Hierarchically Merging and Agent Refinement ROUGE-L 18.62 #3 of 10 Archive leaderboard report
Long-Form Narrative Summarization MovieSum Hierarchically Merging and Agent Refinement BERTScore (F1) 59.32 #2 of 5 Archive leaderboard report
Long-Form Narrative Summarization MovieSum Hierarchically Merging and Agent Refinement ROUGE-1 31.31 #2 of 5 Archive leaderboard report
Long-Form Narrative Summarization MovieSum Hierarchically Merging and Agent Refinement ROUGE-2 8.81 #2 of 5 Archive leaderboard report
Long-Form Narrative Summarization MovieSum Hierarchically Merging and Agent Refinement ROUGE-L 18.62 #2 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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