mlsanityData to 7 Oct 2026

Read Before You Think: Mitigating LLM Comprehension Failures with Step-by-Step Reading

arXiv 2504.09402 ↗First posted 13 Apr 2025cs.CL cs.AIcited by 8

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Large Language Models (LLMs) often fail on complex reasoning tasks due to flawed question comprehension, not just flawed logic. This paper presents a systematic investigation into these comprehension failures. Our work yields three key insights: (1) the step-by-step principle, effective for calculation, can be migrated to the reading process to enhance comprehension; (2) increasing the proportion of question-related tokens (e.g., via repetition) succeeds by refocusing attention, a mechanism that can be explicitly controlled; and (3) backward dependencies represent a core bottleneck for decoder-only models that persists even with strong methods like Chain-of-Thought. Based on these findings, we introduce the Step-by-Step Reading (SSR) family of prompts. This multi-stage approach culminates in SSR++, a method specifically engineered to deepen model comprehension by guiding it to parse questions with finer granularity, focus attention on critical tokens, and resolve backward dependencies through iterative re-contextualization. SSR++ sets a new state-of-the-art on multiple reasoning benchmarks, and our analysis confirms it works by directly mitigating semantic misunderstanding. These results demonstrate that guiding how a model reads is a powerful and efficient method for improving its reasoning ability.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 15 of its 19 references that are in the map. First posted on arXiv 13 Apr 2025.

Builds on

15 of 19 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Re-Reading Improves Reasoning in Large Language Models12 Sep

Other references · 14

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2017Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problemscited by 642
  4. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  5. 2023Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Modelscited by 550
  6. 2021A Diverse Corpus for Evaluating and Developing English Math Word Problem Solverscited by 314
  7. 2023Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselvescited by 123
  8. 2023Adapting Large Language Models to Domains via Reading Comprehensioncited by 48
  9. 2025Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?cited by 43
  10. 2024Insights into LLM Long-Context Failures: When Transformers Know but Don't Tellcited by 36
  11. 2023BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddingscited by 32
  12. 2024Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Better Solvers for Math Word Problemscited by 31
2 more
  1. 2023EchoPrompt: Instructing the Model to Rephrase Queries for Improved In-context Learningcited by 13
  2. 2025Large Language Models in Bioinformatics: A Surveycited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 2.

8 papers.

  1. 2025A Survey of Context Engineering for Large Language Modelscited by 120
  2. 2025ZeroTuning: Unlocking the Initial Token's Power to Enhance Large Language Models Without Trainingcited by 11
  3. 2025Can LLMs Handle WebShell Detection? Overcoming Detection Challenges with Behavioral Function-Aware Frameworkcited by 5
  4. 2025CLAUSE: Agentic Neuro-Symbolic Knowledge Graph Reasoning via Dynamic Learnable Context Engineeringcited by 3
  5. 2026Exposing and Defending the Achilles' Heel of Video Mixture-of-Expertscited by 3
  6. 2025Beyond Detection: A Comprehensive Benchmark and Study on Representation Learning for Fine-Grained Webshell Family Classificationcited by 1
  7. 2025Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Datacited by 0
  8. 2026Ask Twice, Look Twice: Prompt Echoing Resolves the Question-First Paradox in Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansion
  2. 2024MMLU-SR: A Benchmark for Stress-Testing Reasoning Capability of Large Language Models
  3. 2024Hierarchical Deconstruction of LLM Reasoning: A Graph-Based Framework for Analyzing Knowledge Utilization
  4. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  5. 2025Answer-Centric or Reasoning-Driven? Uncovering the Latent Memory Anchor in LLMs
  6. 2025Prompting Test-Time Scaling Is A Strong LLM Reasoning Data Augmentation
  7. 2025Emulating Retrieval Augmented Generation via Prompt Engineering for Enhanced Long Context Comprehension in LLMs
  8. 2024General Purpose Verification for Chain of Thought Prompting