# Contrastive Decoding Improves Reasoning in Large Language Models

arXiv: https://arxiv.org/abs/2309.09117 · HTML: https://mlsanity.com/paper/2309.09117/

## Key facts

- Cited by 74 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 30 of its 35 references that are in the map.
- First posted on arXiv 17 Sep 2023.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 30 of 30)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [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)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [Are NLP Models really able to Solve Simple Math Word Problems?](https://mlsanity.com/paper/2103.07191/) (2021)
- [Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies](https://mlsanity.com/paper/2101.02235/) (2021)
- [Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problems](https://mlsanity.com/paper/1705.04146/) (2017)
- [Contrastive Decoding: Open-ended Text Generation as Optimization](https://mlsanity.com/paper/2210.15097/) (2022)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Models](https://mlsanity.com/paper/2309.03883/) (2023)
- [A Diverse Corpus for Evaluating and Developing English Math Word Problem Solvers](https://mlsanity.com/paper/2106.15772/) (2021)
- [Reasoning with Language Model Prompting: A Survey](https://mlsanity.com/paper/2212.09597/) (2022)
- [Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters](https://mlsanity.com/paper/2212.10001/) (2022)
- [ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoning](https://mlsanity.com/paper/2212.07919/) (2022)
- [Locally Typical Sampling](https://mlsanity.com/paper/2202.00666/) (2022)
- [Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance](https://mlsanity.com/paper/2305.17306/) (2023)
- [Surfacing Biases in Large Language Models using Contrastive Input Decoding](https://mlsanity.com/paper/2305.07378/) (2023)

## Built on by (most cited first, 30 of 74)

- [Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding](https://mlsanity.com/paper/2311.16922/) (2023)
- [Chain-of-Thought Reasoning Without Prompting](https://mlsanity.com/paper/2402.10200/) (2024)
- [Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability](https://mlsanity.com/paper/2411.19943/) (2024)
- [Mitigating Object Hallucinations in Large Vision-Language Models with Assembly of Global and Local Attention](https://mlsanity.com/paper/2406.12718/) (2024)
- [Interpretable Contrastive Monte Carlo Tree Search Reasoning](https://mlsanity.com/paper/2410.01707/) (2024)
- [The Consensus Game: Language Model Generation via Equilibrium Search](https://mlsanity.com/paper/2310.09139/) (2023)
- [mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models](https://mlsanity.com/paper/2406.02301/) (2024)
- [Spotlight Your Instructions: Instruction-following with Dynamic Attention Steering](https://mlsanity.com/paper/2505.12025/) (2025)
- [Supervised Knowledge Makes Large Language Models Better In-context Learners](https://mlsanity.com/paper/2312.15918/) (2023)
- [Seeing the Image: Prioritizing Visual Correlation by Contrastive Alignment](https://mlsanity.com/paper/2405.17871/) (2024)
- [Unchosen Experts Can Contribute Too: Unleashing MoE Models' Power by Self-Contrast](https://mlsanity.com/paper/2405.14507/) (2024)
- [Entropy-Based Decoding for Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2406.17519/) (2024)
- [Distillation Contrastive Decoding: Improving LLMs Reasoning with Contrastive Decoding and Distillation](https://mlsanity.com/paper/2402.14874/) (2024)
- [Search, Verify and Feedback: Towards Next Generation Post-training Paradigm of Foundation Models via Verifier Engineering](https://mlsanity.com/paper/2411.11504/) (2024)
- [Learning From Correctness Without Prompting Makes LLM Efficient Reasoner](https://mlsanity.com/paper/2403.19094/) (2024)
- [Rapid Optimization for Jailbreaking LLMs via Subconscious Exploitation and Echopraxia](https://mlsanity.com/paper/2402.05467/) (2024)
- [Fast and Slow Generating: An Empirical Study on Large and Small Language Models Collaborative Decoding](https://mlsanity.com/paper/2406.12295/) (2024)
- [Pseudo-RIS: Distinctive Pseudo-supervision Generation for Referring Image Segmentation](https://mlsanity.com/paper/2407.07412/) (2024)
- [VaLiD: Mitigating the Hallucination of Large Vision Language Models by Visual Layer Fusion Contrastive Decoding](https://mlsanity.com/paper/2411.15839/) (2024)
- [LLM2: Let Large Language Models Harness System 2 Reasoning](https://mlsanity.com/paper/2412.20372/) (2024)
- [Fast Large Language Model Collaborative Decoding via Speculation](https://mlsanity.com/paper/2502.01662/) (2025)
- [Context-aware Decoding Reduces Hallucination in Query-focused Summarization](https://mlsanity.com/paper/2312.14335/) (2023)
- [TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention](https://mlsanity.com/paper/2503.10602/) (2025)
- [On the Consistency of Multilingual Context Utilization in Retrieval-Augmented Generation](https://mlsanity.com/paper/2504.00597/) (2025)
- [UCD: Unlearning in LLMs via Contrastive Decoding](https://mlsanity.com/paper/2506.12097/) (2025)
- [Towards Building Specialized Generalist AI with System 1 and System 2 Fusion](https://mlsanity.com/paper/2407.08642/) (2024)
- [Improving Large Language Model (LLM) fidelity through context-aware grounding: A systematic approach to reliability and veracity](https://mlsanity.com/paper/2408.04023/) (2024)
- [Integrative Decoding: Improve Factuality via Implicit Self-consistency](https://mlsanity.com/paper/2410.01556/) (2024)
- [UniDetox: Universal Detoxification of Large Language Models via Dataset Distillation](https://mlsanity.com/paper/2504.20500/) (2025)
- [ProxyThinker: Test-Time Guidance through Small Visual Reasoners](https://mlsanity.com/paper/2505.24872/) (2025)

## Abstract (arXiv)

We demonstrate that Contrastive Decoding -- a simple, computationally light, and training-free text generation method proposed by Li et al 2022 -- achieves large out-of-the-box improvements over greedy decoding on a variety of reasoning tasks. Originally shown to improve the perceived quality of long-form text generation, Contrastive Decoding searches for strings that maximize a weighted difference in likelihood between strong and weak models. We show that Contrastive Decoding leads LLaMA-65B to outperform LLaMA 2, GPT-3.5 and PaLM 2-L on the HellaSwag commonsense reasoning benchmark, and to outperform LLaMA 2, GPT-3.5 and PaLM-540B on the GSM8K math word reasoning benchmark, in addition to improvements on a collection of other tasks. Analysis suggests that Contrastive Decoding improves over existing methods by preventing some abstract reasoning errors, as well as by avoiding simpler modes such as copying sections of the input during chain-of-thought. Overall, Contrastive Decoding outperforms nucleus sampling for long-form generation and greedy decoding for reasoning tasks, making it a powerful general purpose method for generating text from language models.