mlsanityData to 14 Oct 2026

Understanding LLM Reasoning for Abstractive Summarization

arXiv 2512.03503 ↗First posted 3 Dec 2025cs.CLcited by 12

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Reasoning has substantially improved Large Language Models (LLMs) on analytical tasks such as mathematics and code generation, but its value for abstractive summarization remains unclear. To address this gap, we adapt general reasoning strategies to the summarization setting and conduct a large-scale comparative study of 8 reasoning strategies and 3 Large Reasoning Models (LRMs) across 8 diverse datasets, evaluating both summary quality and factual faithfulness. Our results show that reasoning is not a universal solution and its effectiveness depends strongly on the strategy and the summarization setting. In particular, we find a trade-off between summary quality and factual faithfulness. Explicit reasoning strategies often improve reference-based quality, but may weaken factual grounding, whereas implicit reasoning in LRMs shows the opposite tendency. We further find that increasing an LRM's internal reasoning budget does not reliably improve summarization and can even reduce factual consistency. These findings suggest that, for summarization, more reasoning is not always better. Effective reasoning should preserve faithful compression rather than induce over-elaboration. Our source code is publicly available.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 19 of its 73 references that are in the map. First posted on arXiv 3 Dec 2025.

Builds on

19 of 73 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Semantic-Aware Logical Reasoning via a Semiotic Framework29 Sep
  2. 2025StrucSum: Graph-Structured Reasoning for Long Document Extractive Summarization with LLMs29 May
  3. 2025Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models16 Jan

Other references · 16

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  3. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  4. 2024OpenAI o1 System Cardcited by 1,244
  5. 2022News Summarization and Evaluation in the Era of GPT-3cited by 327
  6. 2018WikiHow: A Large Scale Text Summarization Datasetcited by 204
  7. 2023Universal Self-Consistency for Large Language Model Generationcited by 203
  8. 2025SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agentscited by 69
  9. 2025Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedbackcited by 32
  10. 2025Scaling Behaviors of LLM Reinforcement Learning Post-Training: An Empirical Study in Mathematical Reasoningcited by 29
  11. 2024AdaComp: Extractive Context Compression with Adaptive Predictor for Retrieval-Augmented Large Language Modelscited by 10
  12. 2025Context-Aware Hierarchical Merging for Long Document Summarizationcited by 9
4 more
  1. 2025Experience Retrieval-Augmentation with Electronic Health Records Enables Accurate Discharge QAcited by 9
  2. 2025ReFeed: Multi-dimensional Summarization Refinement with Reflective Reasoning on Feedbackcited by 4
  3. 2025QA-prompting: Improving Summarization with Large Language Models using Question-Answeringcited by 4
  4. 2024A Structure-aware Generative Model for Biomedical Event Extractioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 12.

12 papers.

  1. 2026TEMA: Anchor the Image, Follow the Text for Multi-Modification Composed Image Retrievalcited by 15
  2. 2026ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratchcited by 10
  3. 2026ATTNPO: Attention-Guided Process Supervision for Efficient Reasoningcited by 9
  4. 2026FoE: Forest of Errors Makes the First Solution the Best in Large Reasoning Modelscited by 8
  5. 2026Stable-RAG: Mitigating Retrieval-Permutation-Induced Hallucinations in Retrieval-Augmented Generationcited by 5
  6. 2026ActorMind: Emulating Human Actor Reasoning for Speech Role-Playingcited by 4
  7. 2026Beyond Output Correctness: Benchmarking and Evaluating Large Language Model Reasoning in Coding Taskscited by 1
  8. 2026OmniTrend: Content-Context Modeling for Scalable Social Popularity Predictioncited by 1
  9. 2026A Large-Scale Multi-Dimensional Empirical Study of LLMs for Conversation Summarizationcited by 1
  10. 2026LLMs Should Incorporate Explicit Mechanisms for Human Empathycited by 0
  11. 2026HAVEN: Hierarchically Aligned Multimodal Benchmark for Unified Video Understandingcited by 0
  12. 2026VoxSumm: A Multilingual Corpus of Long-Form Spoken News for Joint Summarization and Translationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Large Language Models are Not Yet Human-Level Evaluators for Abstractive Summarization
  2. 2025DeepSeek-R1 vs. o3-mini: How Well can Reasoning LLMs Evaluate MT and Summarization?
  3. 2024Meaningful Learning: Enhancing Abstract Reasoning in Large Language Models via Generic Fact Guidance
  4. 2025Reasoning or Not? A Comprehensive Evaluation of Reasoning LLMs for Dialogue Summarization
  5. 2023Evaluating Factual Consistency of Summaries with Large Language Models
  6. 2025Do LLMs Overthink Basic Math Reasoning? Benchmarking the Accuracy-Efficiency Tradeoff in Language Models
  7. 2025Exploring the Hidden Reasoning Process of Large Language Models by Misleading Them
  8. 2023LLMs and the Abstraction and Reasoning Corpus: Successes, Failures, and the Importance of Object-based Representations