Skip to content
← Back to explorer

Lost in Stories: Consistency Bugs in Long Story Generation by LLMs

Junjie Li, Xinrui Guo, Yuhao Wu, Roy Ka-Wei Lee, Hongzhi Li, Yutao Xie · Mar 6, 2026 · Citations: 0

Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Mar 6, 2026, 4:16 AM

Recent

Extraction refreshed

Mar 14, 2026, 6:16 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.25

Abstract

What happens when a storyteller forgets its own story? Large Language Models (LLMs) can now generate narratives spanning tens of thousands of words, but they often fail to maintain consistency throughout. When generating long-form narratives, these models can contradict their own established facts, character traits, and world rules. Existing story generation benchmarks focus mainly on plot quality and fluency, leaving consistency errors largely unexplored. To address this gap, we present ConStory-Bench, a benchmark designed to evaluate narrative consistency in long-form story generation. It contains 2,000 prompts across four task scenarios and defines a taxonomy of five error categories with 19 fine-grained subtypes. We also develop ConStory-Checker, an automated pipeline that detects contradictions and grounds each judgment in explicit textual evidence. Evaluating a range of LLMs through five research questions, we find that consistency errors show clear tendencies: they are most common in factual and temporal dimensions, tend to appear around the middle of narratives, occur in text segments with higher token-level entropy, and certain error types tend to co-occur. These findings can inform future efforts to improve consistency in long-form narrative generation. Our project page is available at https://picrew.github.io/constory-bench.github.io/.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction flags indicate low-signal or possible false-positive protocol mapping.
  • Extraction confidence is 0.25 (below strong-reference threshold).
  • No explicit evaluation mode was extracted from available metadata.
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

Background context only.

Main weakness

Extraction flags indicate low-signal or possible false-positive protocol mapping.

Trust level

Low

Eval-Fit Score

0/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Weak / implicit signal

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Persisted extraction missing

No explicit feedback protocol extracted.

Evidence snippet: What happens when a storyteller forgets its own story?

Evaluation Modes

missing

None explicit

Confidence: Low Source: Persisted extraction missing

Validate eval design from full paper text.

Evidence snippet: What happens when a storyteller forgets its own story?

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: What happens when a storyteller forgets its own story?

Benchmarks / Datasets

partial

Constory Bench

Confidence: Low Source: Persisted extraction evidenced

Useful for quick benchmark comparison.

Evidence snippet: To address this gap, we present ConStory-Bench, a benchmark designed to evaluate narrative consistency in long-form story generation.

Reported Metrics

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No metric anchors detected.

Evidence snippet: What happens when a storyteller forgets its own story?

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: What happens when a storyteller forgets its own story?

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: General
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes:
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.25
  • Flags: low_signal, possible_false_positive

Protocol And Measurement Signals

Benchmarks / Datasets

Constory-Bench

Reported Metrics

No metric terms were extracted from the available abstract.

Research Brief

Deterministic synthesis

Existing story generation benchmarks focus mainly on plot quality and fluency, leaving consistency errors largely unexplored. HFEPX protocol signal is limited in abstract-level metadata, so treat it as adjacent context. Updated from current HFEPX corpus.

Generated Mar 14, 2026, 6:16 AM · Grounded in abstract + metadata only

Key Takeaways

  • Existing story generation benchmarks focus mainly on plot quality and fluency, leaving consistency errors largely unexplored.
  • To address this gap, we present ConStory-Bench, a benchmark designed to evaluate narrative consistency in long-form story generation.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Cross-check benchmark overlap: Constory-Bench.
  • Verify metric definitions before comparing against your eval pipeline.

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Low-signal flag detected: protocol relevance may be indirect.

Research Summary

Contribution Summary

  • Existing story generation benchmarks focus mainly on plot quality and fluency, leaving consistency errors largely unexplored.
  • To address this gap, we present ConStory-Bench, a benchmark designed to evaluate narrative consistency in long-form story generation.

Why It Matters For Eval

  • Existing story generation benchmarks focus mainly on plot quality and fluency, leaving consistency errors largely unexplored.
  • To address this gap, we present ConStory-Bench, a benchmark designed to evaluate narrative consistency in long-form story generation.

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Gap: Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Pass: Benchmark or dataset anchors are present

    Detected: Constory-Bench

  • Gap: Metric reporting is present

    No metric terms extracted.

Category-Adjacent Papers (Broader Context)

These papers are nearby in arXiv category and useful for broader context, but not necessarily protocol-matched to this paper.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote