mlsanityData to 7 Oct 2026

Learning Task Decomposition to Assist Humans in Competitive Programming

arXiv 2406.04604 ↗First posted 7 Jun 2024cs.CL cs.PLcited by 14

Not on the map. Why?

Abstract (arXiv)

When using language models (LMs) to solve complex problems, humans might struggle to understand the LM-generated solutions and repair the flawed ones. To assist humans in repairing them, we propose to automatically decompose complex solutions into multiple simpler pieces that correspond to specific subtasks. We introduce a novel objective for learning task decomposition, termed assistive value (AssistV), which measures the feasibility and speed for humans to repair the decomposed solution. We collect a dataset of human repair experiences on different decomposed solutions. Utilizing the collected data as in-context examples, we then learn to critique, refine, and rank decomposed solutions to improve AssistV. We validate our method under competitive programming problems: under 177 hours of human study, our method enables non-experts to solve 33.3% more problems, speeds them up by 3.3x, and empowers them to match unassisted experts.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Builds on 23 of its 41 references that are in the map. First posted on arXiv 7 Jun 2024.

Builds on

23 of 41 references.

References · 23

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  3. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  4. 2021Program Synthesis with Large Language Modelscited by 3,509
  5. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  6. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  7. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  8. 2023Code Llama: Open Foundation Models for Codecited by 2,105
  9. 2016Concrete Problems in AI Safetycited by 1,582
  10. 2022Competition-Level Code Generation with AlphaCodecited by 1,427
  11. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 965
  12. 2023Teaching Large Language Models to Self-Debugcited by 963
11 more
  1. 2021Measuring Coding Challenge Competence With APPScited by 879
  2. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  3. 2022Self-critiquing models for assisting human evaluatorscited by 372
  4. 2023Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervisioncited by 318
  5. 2021Recursively Summarizing Books with Human Feedbackcited by 287
  6. 2018Supervising strong learners by amplifying weak expertscited by 139
  7. 2022Successive Prompting for Decomposing Complex Questionscited by 94
  8. 2022Repair Is Nearly Generation: Multilingual Program Repair with LLMscited by 68
  9. 2022Parsel: Algorithmic Reasoning with Language Models by Composing Decompositionscited by 62
  10. 2022Humans decompose tasks by trading off utility and computational costcited by 18
  11. 2022Two-Turn Debate Doesn't Help Humans Answer Hard Reading Comprehension Questionscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 6; 2025: 6; 2026: 2.

14 papers.

Sort:

By area:

  1. 2024Language Models Learn to Mislead Humans via RLHFcited by 118
  2. 2024Prover-Verifier Games improve legibility of LLM outputscited by 79
  3. 2025Code to Think, Think to Code: A Survey on Code-Enhanced Reasoning and Reasoning-Driven Code Intelligence in LLMscited by 47
  4. 2024From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debuggingcited by 46
  5. 2024Tree-of-Code: A Tree-Structured Exploring Framework for End-to-End Code Generation and Execution in Complex Task Handlingcited by 11
  6. 2024The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignmentcited by 11
  7. 2025Unsupervised Elicitation of Language Modelscited by 11
  8. 2025A Good Plan is Hard to Find: Aligning Models with Preferences is Misaligned with What Helps Userscited by 6
  9. 2025AdaptBot: Combining LLM with Knowledge Graphs and Human Input for Generic-to-Specific Task Decomposition and Knowledge Refinementcited by 3
  10. 2024The Superalignment of Superhuman Intelligence with Large Language Modelscited by 2
  11. 2026MT-OSC: Path for LLMs that Get Lost in Multi-Turn Conversationcited by 2
  12. 2025Research Superalignment Should Advance Now with Alternating Competence and Conformity Optimizationcited by 1
2 more
  1. 2025Superalignment with Dynamic Human Valuescited by 1
  2. 2026Towards Neuro-symbolic Causal Rule Synthesis, Verification, and Evaluation Grounded in Legal and Safety Principlescited by 1