mlsanityData to 14 Oct 2026

Many-Shot CoT-ICL: Making In-Context Learning Truly Learn

arXiv 2605.13511 ↗First posted 13 May 2026cs.CL cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

While many-shot ICL achieves remarkable performance, prior studies of its scaling behavior have mainly focused on non-reasoning tasks. In this work, we study many-shot ICL on reasoning tasks, with a particular focus on many-shot chain-of-thought in-context learning (CoT-ICL). Analyzing across non-reasoning and reasoning tasks and across non-reasoning and reasoning-oriented LLMs, we identify several distinctive properties of many-shot CoT-ICL. We further interpret these findings by viewing many-shot CoT-ICL as in-context test-time learning rather than scaled pattern matching, and suggest two principles: (i) demonstrations should be easy for the target model to understand, and (ii) they should be ordered to support a smooth conceptual progression. Guided by the principle, we propose Curvilinear Demonstration Selection (CDS), a simple ordering method that yields up to a 5.42 percentage-point gain on a math task with 64 demonstrations. Overall, our results reframe the long context window from a retrieval buffer into a structured curriculum for in-context test-time learning.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 32 of its 42 references that are in the map. First posted on arXiv 13 May 2026.

Builds on

32 of 42 references.

References · 32

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  5. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  6. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  7. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  8. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  9. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  10. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
  11. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  12. 2022Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Taskscited by 1,076
20 more
  1. 2022Transformers learn in-context by gradient descentcited by 732
  2. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 688
  3. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 601
  4. 2020Efficient Intent Detection with Dual Sentence Encoderscited by 433
  5. 2025rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinkingcited by 313
  6. 2024Long-context LLMs Struggle with Long In-context Learningcited by 286
  7. 2023The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learningcited by 257
  8. 2019Benchmarking Natural Language Understanding Services for building Conversational Agentscited by 166
  9. 2022Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Orderingcited by 161
  10. 2024In-Context Learning with Long-Context Models: An In-Depth Explorationcited by 128
  11. 2023LM-Infinite: Zero-Shot Extreme Length Generalization for Large Language Modelscited by 124
  12. 2022An Information-theoretic Approach to Prompt Engineering Without Ground Truth Labelscited by 89
  13. 2024Induction Heads as an Essential Mechanism for Pattern Matching in In-context Learningcited by 46
  14. 2025Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedbackcited by 41
  15. 2025From Few to Many: Self-Improving Many-Shot Reasoners Through Iterative Optimization and Generationcited by 21
  16. 2025Exploring the Role of Diversity in Example Selection for In-Context Learningcited by 12
  17. 2025The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understandingcited by 11
  18. 2024Selection-p: Self-Supervised Task-Agnostic Prompt Compression for Faithfulness and Transferabilitycited by 9
  19. 2024Revisiting In-Context Learning with Long Context Language Modelscited by 8
  20. 2025PRELUDE: A Benchmark Designed to Require Global Comprehension and Reasoning over Long Contextscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026A New Role for Relevance: Guiding Corpus Interaction in Agentic Searchcited by 3
  2. 2026Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructurecited by 0
  3. 2026InsightEmb: Learning Action-Intent Embeddings for Agentic Insight Retrievalcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Demonstrations, CoT, and Prompting: A Theoretical Analysis of ICL
  2. 2025Revisiting Chain-of-Thought Prompting: Zero-shot Can Be Stronger than Few-shot
  3. 2023In-Context Learning with Iterative Demonstration Selection
  4. 2024What Makes In-context Learning Effective for Mathematical Reasoning: A Theoretical Analysis
  5. 2025To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive CoT Examples
  6. 2024Many-Shot In-Context Learning
  7. 2026Process In-Context Learning: Enhancing Mathematical Reasoning via Dynamic Demonstration Insertion
  8. 2023Dr.ICL: Demonstration-Retrieved In-context Learning