# Customizing Language Model Responses with Contrastive In-Context Learning

arXiv: https://arxiv.org/abs/2401.17390 · HTML: https://mlsanity.com/paper/2401.17390/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 15 references that are in the map.
- First posted on arXiv 30 Jan 2024.

## Builds on (most cited first, 13 of 13)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [Large Language Models Are Human-Level Prompt Engineers](https://mlsanity.com/paper/2211.01910/) (2022)
- [Selective Annotation Makes Language Models Better Few-Shot Learners](https://mlsanity.com/paper/2209.01975/) (2022)
- [Instruction Induction: From Few Examples to Natural Language Task Descriptions](https://mlsanity.com/paper/2205.10782/) (2022)
- [COCO-LM: Correcting and Contrasting Text Sequences for Language Model Pretraining](https://mlsanity.com/paper/2102.08473/) (2021)
- [Dialogue Response Ranking Training with Large-Scale Human Feedback Data](https://mlsanity.com/paper/2009.06978/) (2020)
- [AutoHint: Automatic Prompt Optimization with Hint Generation](https://mlsanity.com/paper/2307.07415/) (2023)

## Built on by (most cited first, 19 of 19)

- [Measuring What Makes You Unique: Difference-Aware User Modeling for Enhancing LLM Personalization](https://mlsanity.com/paper/2503.02450/) (2025)
- [Q-Adapter: Customizing Pre-trained LLMs to New Preferences with Forgetting Mitigation](https://mlsanity.com/paper/2407.03856/) (2024)
- [Does Few-Shot Learning Help LLM Performance in Code Synthesis?](https://mlsanity.com/paper/2412.02906/) (2024)
- [PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling](https://mlsanity.com/paper/2502.01925/) (2025)
- [LLM-Meta-SR: In-Context Learning for Evolving Selection Operators in Symbolic Regression](https://mlsanity.com/paper/2505.18602/) (2025)
- [Network-Level Prompt and Trait Leakage in Local Research Agents](https://mlsanity.com/paper/2508.20282/) (2025)
- [Learning to Search Effective Example Sequences for In-Context Learning](https://mlsanity.com/paper/2503.08030/) (2025)
- [CG-TTRL: Context-Guided Test-Time Reinforcement Learning for On-Device Large Language Models](https://mlsanity.com/paper/2511.06430/) (2025)
- [RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning](https://mlsanity.com/paper/2601.00086/) (2025)
- [A Survey of the State-of-the-Art in Conversational Question Answering Systems](https://mlsanity.com/paper/2509.05716/) (2025)
- [LLM For Loop Invariant Generation and Fixing: How Far Are We?](https://mlsanity.com/paper/2511.06552/) (2025)
- [Seeing the Undefined: Chain-of-Action for Generative Semantic Labels](https://mlsanity.com/paper/2411.17406/) (2024)
- [Comparative Personalization for Multi-document Summarization](https://mlsanity.com/paper/2509.21562/) (2025)
- [Empathy Applicability Modeling for General Health Queries](https://mlsanity.com/paper/2601.09696/) (2026)
- [A Training-Free Regeneration Paradigm: Contrastive Reflection Memory Guided Self-Verification and Self-Improvement](https://mlsanity.com/paper/2603.20441/) (2026)
- [Supplement Generation Training for Enhancing Agentic Task Performance](https://mlsanity.com/paper/2604.20727/) (2026)
- [ProjAgent: Procedural Similarity Retrieval for Repository-Level Code Generation](https://mlsanity.com/paper/2607.08691/) (2026)
- [Smarter by the Moment: Environment-Driven Dynamic Policies for Continual LLM Improvement](https://mlsanity.com/paper/2609.16800/) (2026)
- [The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale](https://mlsanity.com/paper/2609.26087/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are becoming increasingly important for machine learning applications. However, it can be challenging to align LLMs with our intent, particularly when we want to generate content that is preferable over others or when we want the LLM to respond in a certain style or tone that is hard to describe. To address this challenge, we propose an approach that uses contrastive examples to better describe our intent. This involves providing positive examples that illustrate the true intent, along with negative examples that show what characteristics we want LLMs to avoid. The negative examples can be retrieved from labeled data, written by a human, or generated by the LLM itself. Before generating an answer, we ask the model to analyze the examples to teach itself what to avoid. This reasoning step provides the model with the appropriate articulation of the user's need and guides it towards generting a better answer. We tested our approach on both synthesized and real-world datasets, including StackExchange and Reddit, and found that it significantly improves performance compared to standard few-shot prompting