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

arXiv: https://arxiv.org/abs/2504.09402 · HTML: https://mlsanity.com/paper/2504.09402/

## Key facts

- Cited by 10 papers in the map (data through 14 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.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 15 of 15)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems](https://mlsanity.com/paper/1705.04146/) (2017)
- [Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models](https://mlsanity.com/paper/2305.04091/) (2023)
- [Decomposed Prompting: A Modular Approach for Solving Complex Tasks](https://mlsanity.com/paper/2210.02406/) (2022)
- [A Diverse Corpus for Evaluating and Developing English Math Word Problem Solvers](https://mlsanity.com/paper/2106.15772/) (2021)
- [Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves](https://mlsanity.com/paper/2311.04205/) (2023)
- [Adapting Large Language Models to Domains via Reading Comprehension](https://mlsanity.com/paper/2309.09530/) (2023)
- [Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?](https://mlsanity.com/paper/2502.19361/) (2025)
- [Re-Reading Improves Reasoning in Large Language Models](https://mlsanity.com/paper/2309.06275/) (2023)
- [Insights into LLM Long-Context Failures: When Transformers Know but Don't Tell](https://mlsanity.com/paper/2406.14673/) (2024)
- [BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddings](https://mlsanity.com/paper/2311.05296/) (2023)
- [Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Better Solvers for Math Word Problems](https://mlsanity.com/paper/2404.14963/) (2024)
- [EchoPrompt: Instructing the Model to Rephrase Queries for Improved In-context Learning](https://mlsanity.com/paper/2309.10687/) (2023)
- [Large Language Models in Bioinformatics: A Survey](https://mlsanity.com/paper/2503.04490/) (2025)

## Built on by (most cited first, 10 of 10)

- [A Survey of Context Engineering for Large Language Models](https://mlsanity.com/paper/2507.13334/) (2025)
- [ZeroTuning: Unlocking the Initial Token's Power to Enhance Large Language Models Without Training](https://mlsanity.com/paper/2505.11739/) (2025)
- [Can LLMs Handle WebShell Detection? Overcoming Detection Challenges with Behavioral Function-Aware Framework](https://mlsanity.com/paper/2504.13811/) (2025)
- [CLAUSE: Agentic Neuro-Symbolic Knowledge Graph Reasoning via Dynamic Learnable Context Engineering](https://mlsanity.com/paper/2509.21035/) (2025)
- [Exposing and Defending the Achilles' Heel of Video Mixture-of-Experts](https://mlsanity.com/paper/2602.01369/) (2026)
- [Beyond Detection: A Comprehensive Benchmark and Study on Representation Learning for Fine-Grained Webshell Family Classification](https://mlsanity.com/paper/2512.05288/) (2025)
- [Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Data](https://mlsanity.com/paper/2511.19498/) (2025)
- [Ask Twice, Look Twice: Prompt Echoing Resolves the Question-First Paradox in Vision-Language Models](https://mlsanity.com/paper/2607.15565/) (2026)
- [Second-Order Problem Solving for Recursive Self-Improvement in Formal Verification](https://mlsanity.com/paper/2610.05701/) (2026)
- [What Breaks in Multi-Image Schema Application? Factorial Oracle Interventions Reveal an Instance-Information Bottleneck](https://mlsanity.com/paper/2610.19611/) (2026)

## 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.