# Structural Anchors and Reasoning Fragility:Understanding CoT Robustness in LLM4Code

arXiv: https://arxiv.org/abs/2604.12214 · HTML: https://mlsanity.com/paper/2604.12214/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 20 of its 51 references that are in the map.
- First posted on arXiv 14 Apr 2026.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 20 of 20)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [Measuring Coding Challenge Competence With APPS](https://mlsanity.com/paper/2105.09938/) (2021)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation](https://mlsanity.com/paper/2302.09664/) (2023)
- [GraphCodeBERT: Pre-training Code Representations with Data Flow](https://mlsanity.com/paper/2009.08366/) (2020)
- [BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions](https://mlsanity.com/paper/2406.15877/) (2024)
- [Chain of Code: Reasoning with a Language Model-Augmented Code Emulator](https://mlsanity.com/paper/2312.04474/) (2023)
- [ReCode: Robustness Evaluation of Code Generation Models](https://mlsanity.com/paper/2212.10264/) (2022)
- [Instruction Tuning for Secure Code Generation](https://mlsanity.com/paper/2402.09497/) (2024)
- [Towards Better Chain-of-Thought Prompting Strategies: A Survey](https://mlsanity.com/paper/2310.04959/) (2023)
- [Uncertainty-Guided Chain-of-Thought for Code Generation with LLMs](https://mlsanity.com/paper/2503.15341/) (2025)
- [MHPP: Exploring the Capabilities and Limitations of Language Models Beyond Basic Code Generation](https://mlsanity.com/paper/2405.11430/) (2024)
- [An Empirical Study on the Code Refactoring Capability of Large Language Models](https://mlsanity.com/paper/2411.02320/) (2024)
- [THINK-Bench: Evaluating Thinking Efficiency and Chain-of-Thought Quality of Large Reasoning Models](https://mlsanity.com/paper/2505.22113/) (2025)

## Abstract (arXiv)

Chain-of-Thought (CoT) prompting is widely used to elicit explicit reasoning from large language models for code (LLM4Code). However, its impact on robustness and the stability of reasoning trajectories under realistic input perturbations remains poorly understood. Prior work has largely evaluated CoT through final correctness, leaving a critical gap in understanding how CoT reshapes internal uncertainty dynamics and why it sometimes harms rather than helps code generation. We suggest that CoT is not uniformly beneficial; instead, its robustness depends on whether perturbations destabilize structurally sensitive commitment points along the reasoning-to-code trajectory. We conduct a controlled, large-scale empirical study of CoT across six models and two code benchmarks (MHPP and BigCodeBench), subjecting task docstrings to systematic character-, word-, and sentence-level perturbations. We instrument full generation traces with token-level uncertainty and define three novel structural anchors: reasoning-code transition, symbolic commitment, and algorithmic articulation. Findings: (1) CoT does not yield uniform performance or robustness gains: its benefits are contingent on model family, task structure, and prompt explicitness. (2) CoT and No-CoT exhibit distinct robustness profiles, with different perturbation families triggering different failure modes. (3) We identify three recurrent trajectory deformations--Lengthening, Branching, and Simplification--that systematically emerge when perturbations interact with structural anchors and explain failure patterns. (4) Early-stage uncertainty serves as a reliable diagnostic signal for localizing where trajectory instability begins around sensitive anchors. These results provide a unified explanation for CoT's mixed performance and suggest design principles for building more robust reasoning-based code generators.