mlsanityData to 7 Oct 2026

A Self Consistency Based Reranking for Narrative Question Answering

arXiv 2606.15741 ↗First posted 14 Jun 2026cs.CL cs.AIcited by 0

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Narrative question answering (NQA) is a challenging task in natural language processing that requires models to understand long textual contexts, capture relationships across events, and generate coherent responses. Despite recent advances in pretrained language models, most existing approaches rely on a single decoding output during inference, making them sensitive to generation variability and often resulting in incomplete or inconsistent answers .To address this limitation, we propose a self-ensemble Self-Consistency-Based reranking framework for narrative question answering. The proposed method generates multiple candidate answers for each story-question pair and selects the final answer based on semantic agreement among the generated responses. This allows the model to explore diverse answer formulations while improving robustness through consensus-based selection without requiring modifications to the underlying architecture .The framework combines pretrained and fine-tuned language generation with multi-answer inference and similarity-based reranking. We evaluate the proposed approach on the NarrativeQA dataset using multiple models, including FLAN-T5 (Base and Small) and Pegasus-Large, under both baseline and fine-tuned settings .Experimental results demonstrate that the proposed method consistently improves performance across all models. In particular, FLAN-T5-Base achieves the best overall performance, improving from 82.32% to 86.66% (+4.34%) when combined with self-ensemble inference. Additionally, the largest improvement is observed with Pegasus-Large, which increases from 72.50% to 87.07% (+14.57%), highlighting the effectiveness of the proposed strategy.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 25 of its 30 references that are in the map. First posted on arXiv 14 Jun 2026.

Builds on

25 of 30 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning17 Oct
  2. 2025Revisiting Self-Consistency from Dynamic Distributional Alignment Perspective on Answer Aggregation27 Feb
  3. 2025Bridging Internal Probability and Self-Consistency for Effective and Efficient LLM Reasoning1 Feb
  4. 2024Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling30 Aug
  5. 2023Universal Self-Consistency for Large Language Model Generation29 Nov

Other references · 20

  1. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  2. 2023Let's Verify Step by Stepcited by 3,598
  3. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,168
  4. 2023Better patching using LLM prompting, via Self-Consistencycited by 29
  5. 2024Large Language Model (LLM) AI text generation detection based on transformer deep learning algorithmcited by 23
  6. 2020Importance of Self-Consistency in Active Learning for Semantic Segmentationcited by 21
  7. 2024MedSyn: LLM-based Synthetic Medical Text Generation Frameworkcited by 19
  8. 2025Self-Critique and Refinement for Faithful Natural Language Explanationscited by 9
  9. 2024LLM-DaaS: LLM-driven Drone-as-a-Service Operations from Text User Requestscited by 7
  10. 2025Self-Improvement of Language Models by Post-Training on Multi-Agent Debatecited by 7
  11. 2024LLM-SEM: A Sentiment-Based Student Engagement Metric Using LLMS for E-Learning Platformscited by 6
  12. 2024RIRO: Reshaping Inputs, Refining Outputs Unlocking the Potential of Large Language Models in Data-Scarce Contextscited by 6
8 more
  1. 2024ERPA: Efficient RPA Model Integrating OCR and LLMs for Intelligent Document Processingcited by 6
  2. 2024Beyond Self-Consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Stagingcited by 4
  3. 2025An Ensemble Classification Approach in A Multi-Layered Large Language Model Framework for Disease Predictioncited by 4
  4. 2021Scalable Consistency Training for Graph Neural Networks via Self-Ensemble Self-Distillationcited by 2
  5. 2025Two-Stage Quranic QA via Ensemble Retrieval and Instruction-Tuned Answer Extractioncited by 2
  6. 2026Severity-Aware Weighted Loss for Arabic Medical Text Generationcited by 2
  7. 2026A Severity-Based Curriculum Learning Strategy for Arabic Medical Text Generationcited by 2
  8. 2025Confidence-Credibility Aware Weighted Ensembles of Small LLMs Outperform Large LLMs in Emotion Detectioncited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Boosting Self-Consistency with Ranking
  2. 2025Chronological Passage Assembling in RAG framework for Temporal Question Answering
  3. 2024NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens
  4. 2023Self-Consistent Narrative Prompts on Abductive Natural Language Inference
  5. 2025NovelHopQA: Diagnosing Multi-Hop Reasoning Failures in Long Narrative Contexts
  6. 2025Can we repurpose multiple-choice question-answering models to rerank retrieved documents? ↗
  7. 2025Self-Taught Agentic Long Context Understanding
  8. 2026Verbal-R3: Verbal Reranker as the Missing Bridge between Retrieval and Reasoning