# Why Thinking Hurts: Diagnosing and Rectifying Linguistic Inertia in Large Language Models for Recommendation

arXiv: https://arxiv.org/abs/2602.16587 · HTML: https://mlsanity.com/paper/2602.16587/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning.
- Builds on 19 of its 45 references that are in the map.
- First posted on arXiv 18 Feb 2026.

## Placed in

- LLM reasoning / [LLM reasoning for recommendation](https://mlsanity.com/lr/llm-reasoning-for-recommendation/)

## Builds on (most cited first, 19 of 19)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Chat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender System](https://mlsanity.com/paper/2303.14524/) (2023)
- [Is ChatGPT a Good Recommender? A Preliminary Study](https://mlsanity.com/paper/2304.10149/) (2023)
- [Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendations](https://mlsanity.com/paper/2402.17152/) (2024)
- [Think Before Recommend: Unleashing the Latent Reasoning Power for Sequential Recommendation](https://mlsanity.com/paper/2503.22675/) (2025)
- [OneRec-Think: In-Text Reasoning for Generative Recommendation](https://mlsanity.com/paper/2510.11639/) (2025)
- [Review-driven Personalized Preference Reasoning with Large Language Models for Recommendation](https://mlsanity.com/paper/2408.06276/) (2024)
- [Reason4Rec: Deliberative User Preference Alignment of Large Language Models for Recommendation](https://mlsanity.com/paper/2502.02061/) (2025)
- [Thought-Augmented Planning for LLM-Powered Interactive Recommender Agent](https://mlsanity.com/paper/2506.23485/) (2025)
- [Pearl: A Review-driven Persona-Knowledge Grounded Conversational Recommendation Dataset](https://mlsanity.com/paper/2403.04460/) (2024)
- [OpenOneRec Technical Report](https://mlsanity.com/paper/2512.24762/) (2025)
- [Scaling New Frontiers: Insights into Large Recommendation Models](https://mlsanity.com/paper/2412.00714/) (2024)
- [Slow Thinking for Sequential Recommendation](https://mlsanity.com/paper/2504.09627/) (2025)
- [The Next Paradigm Is User-Centric Agent, Not Platform-Centric Service](https://mlsanity.com/paper/2602.15682/) (2026)
- [FuXi-β: Towards a Lightweight and Fast Large-Scale Generative Recommendation Model](https://mlsanity.com/paper/2508.10615/) (2025)
- [Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control](https://mlsanity.com/paper/2602.15659/) (2026)
- [HyMiRec: A Hybrid Multi-interest Learning Framework for LLM-based Sequential Recommendation](https://mlsanity.com/paper/2510.13738/) (2025)
- [Revisiting scalable sequential recommendation with Multi-Embedding Approach and Mixture-of-Experts](https://mlsanity.com/paper/2510.25285/) (2025)

## Built on by (most cited first, 9 of 9)

- [OneReason Technical Report](https://mlsanity.com/paper/2606.06260/) (2026)
- [Learning from Emptiness: De-biasing Listwise Rerankers with Content-Agnostic Probability Calibration](https://mlsanity.com/paper/2604.10150/) (2026)
- [Efficient Personalized Reranking with Semi-Autoregressive Generation and Online Knowledge Distillation](https://mlsanity.com/paper/2603.07107/) (2026)
- [IE as Cache: Information Extraction Enhanced Agentic Reasoning](https://mlsanity.com/paper/2604.14930/) (2026)
- [SAPO: Step-Aligned Policy Optimization for Reasoning-Based Generative Recommendation](https://mlsanity.com/paper/2605.17648/) (2026)
- [AsymSpec: Context-Asymmetric Speculative Decoding for Agentic LLMs](https://mlsanity.com/paper/2608.26004/) (2026)
- [Rethinking the Necessity of Adaptive Retrieval-Augmented Generation through the Lens of Adaptive Listwise Ranking](https://mlsanity.com/paper/2604.15621/) (2026)
- [WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models](https://mlsanity.com/paper/2607.26621/) (2026)
- [The Disconnect Between Better Descriptive Reasoning Trace Quality and Recommendation Effectiveness](https://mlsanity.com/paper/2608.23154/) (2026)

## Abstract (arXiv)

Chain-of-Thought (CoT) reasoning is widely used to improve LLM performance, and recent foundation recommender models adopt it by generating textual reasoning before predicting target items represented by Semantic IDs (SIDs). However, we observe that enabling thinking mode in models such as OpenOneRec can degrade recommendation quality by up to 25%. We investigate this failure and identify Linguistic Inertia: when a textual CoT segment is inserted before SID generation, the model relies more on natural-language context and less on historical SID evidence. Further analyses show that this effect is amplified by reduced access to historical information and longer CoT lengths. To mitigate it, we propose Linguistic-Inertia-Calibrated Decoding (LICD), a training-free framework that combines Reasoning-Chain Compression and Bias-Subtracted Contrastive Inference. Experiments on three large-scale benchmarks show that LICD consistently outperforms both no-thinking and original-thinking baselines. Our code is available at https://github.com/USTC-StarTeam/LICD.