# Self-DC: When to Reason and When to Act? Self Divide-and-Conquer for Compositional Unknown Questions

arXiv: https://arxiv.org/abs/2402.13514 · HTML: https://mlsanity.com/paper/2402.13514/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 26 of its 33 references that are in the map.
- First posted on arXiv 21 Feb 2024.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 26 of 26)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (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)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Measuring and Narrowing the Compositionality Gap in Language Models](https://mlsanity.com/paper/2210.03350/) (2022)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions](https://mlsanity.com/paper/2212.10509/) (2022)
- [A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity](https://mlsanity.com/paper/2302.04023/) (2023)
- [Teaching Models to Express Their Uncertainty in Words](https://mlsanity.com/paper/2205.14334/) (2022)
- [Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy](https://mlsanity.com/paper/2305.15294/) (2023)
- [Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP](https://mlsanity.com/paper/2212.14024/) (2022)
- [FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation](https://mlsanity.com/paper/2310.03214/) (2023)
- [Generate rather than Retrieve: Large Language Models are Strong Context Generators](https://mlsanity.com/paper/2209.10063/) (2022)
- [ToolQA: A Dataset for LLM Question Answering with External Tools](https://mlsanity.com/paper/2306.13304/) (2023)
- [Do Large Language Models Know What They Don't Know?](https://mlsanity.com/paper/2305.18153/) (2023)
- [Reducing conversational agents' overconfidence through linguistic calibration](https://mlsanity.com/paper/2012.14983/) (2020)
- [A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation](https://mlsanity.com/paper/2307.03987/) (2023)
- [Investigating the Factual Knowledge Boundary of Large Language Models with Retrieval Augmentation](https://mlsanity.com/paper/2307.11019/) (2023)
- [Fact-Checking Complex Claims with Program-Guided Reasoning](https://mlsanity.com/paper/2305.12744/) (2023)
- [Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysis](https://mlsanity.com/paper/2210.04714/) (2022)
- [Improving Language Models via Plug-and-Play Retrieval Feedback](https://mlsanity.com/paper/2305.14002/) (2023)
- [Self-Knowledge Guided Retrieval Augmentation for Large Language Models](https://mlsanity.com/paper/2310.05002/) (2023)
- [Cue-CoT: Chain-of-thought Prompting for Responding to In-depth Dialogue Questions with LLMs](https://mlsanity.com/paper/2305.11792/) (2023)

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

- [Knowledge Conflicts for LLMs: A Survey](https://mlsanity.com/paper/2403.08319/) (2024)
- [SMART: Self-Aware Agent for Tool Overuse Mitigation](https://mlsanity.com/paper/2502.11435/) (2025)
- [ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration](https://mlsanity.com/paper/2511.21689/) (2025)
- [AppBench: Planning of Multiple APIs from Various APPs for Complex User Instruction](https://mlsanity.com/paper/2410.19743/) (2024)
- [Position: Agent Should Invoke External Tools ONLY When Epistemically Necessary](https://mlsanity.com/paper/2506.00886/) (2025)
- [Automating Steering for Safe Multimodal Large Language Models](https://mlsanity.com/paper/2507.13255/) (2025)
- [Calibrate-Then-Act: Cost-Aware Exploration in LLM Agents](https://mlsanity.com/paper/2602.16699/) (2026)
- [Query-Level Uncertainty in Large Language Models](https://mlsanity.com/paper/2506.09669/) (2025)
- [CtrlA: Adaptive Retrieval-Augmented Generation via Inherent Control](https://mlsanity.com/paper/2405.18727/) (2024)
- [UAlign: Leveraging Uncertainty Estimations for Factuality Alignment on Large Language Models](https://mlsanity.com/paper/2412.11803/) (2024)
- [Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges](https://mlsanity.com/paper/2505.13328/) (2025)
- [Assessing "Implicit" Retrieval Robustness of Large Language Models](https://mlsanity.com/paper/2406.18134/) (2024)
- [A Desideratum for Conversational Agents: Capabilities, Challenges, and Future Directions](https://mlsanity.com/paper/2504.16939/) (2025)
- [The Quest for Efficient Reasoning: A Data-Centric Benchmark to CoT Distillation](https://mlsanity.com/paper/2505.18759/) (2025)
- [Unified Active Retrieval for Retrieval Augmented Generation](https://mlsanity.com/paper/2406.12534/) (2024)
- [Agentic Knowledgeable Self-awareness](https://mlsanity.com/paper/2504.03553/) (2025)
- [ToolSpectrum : Towards Personalized Tool Utilization for Large Language Models](https://mlsanity.com/paper/2505.13176/) (2025)
- [The Tool-Overuse Illusion: Why Does LLM Prefer External Tools over Internal Knowledge?](https://mlsanity.com/paper/2604.19749/) (2026)
- [SeRTS: Self-Rewarding Tree Search for Biomedical Retrieval-Augmented Generation](https://mlsanity.com/paper/2406.11258/) (2024)
- [TreePS-RAG: Tree-based Process Supervision for Reinforcement Learning in Agentic RAG](https://mlsanity.com/paper/2601.06922/) (2026)
- [To Retrieve or Not to Retrieve? Uncertainty Detection for Dynamic Retrieval Augmented Generation](https://mlsanity.com/paper/2501.09292/) (2025)
- [DAST: Difficulty-Aware Self-Training on Large Language Models](https://mlsanity.com/paper/2503.09029/) (2025)
- [From Manipulation to Mistrust: Explaining Diverse Micro-Video Misinformation for Robust Debunking in the Wild](https://mlsanity.com/paper/2603.25423/) (2026)
- [When to Trust Tools? Adaptive Tool Trust Calibration For Tool-Integrated Math Reasoning](https://mlsanity.com/paper/2604.08281/) (2026)
- [Controlling Tool Use with Heading-Specific Activation Steering](https://mlsanity.com/paper/2607.05790/) (2026)
- [Review-Then-Refine: A Dynamic Framework for Multi-Hop Question Answering with Temporal Adaptability](https://mlsanity.com/paper/2412.15101/) (2024)
- [SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating](https://mlsanity.com/paper/2606.07074/) (2026)
- [MlingConf: A Comprehensive Study of Multilingual Confidence Estimation on Large Language Models](https://mlsanity.com/paper/2410.12478/) (2024)
- [LightSearcher: Efficient DeepSearch via Experiential Memory](https://mlsanity.com/paper/2512.06653/) (2025)
- [Dictionary-Aligned Concept Control for Safeguarding Multimodal LLMs](https://mlsanity.com/paper/2604.08846/) (2026)

## Abstract (arXiv)

Previous research has typically concentrated on leveraging the internal knowledge of Large Language Models (LLMs) to answer known questions (i.e., textitinternal reasoning such as generate-then-read). In contrast, for questions that fall outside their known scope, these models rely on external knowledge retrieval to provide accurate responses (i.e., textitexternal acting such as retrieve-then-read). However, few previous works consider the textitcompositional questions, which consist of several known and unknown sub-questions, necessitating the dynamic combination of previous two methods (i.e., textitinternal reasoning and external acting) to achieve a better trade-off between effectiveness and efficiency. To this end, we introduce a Self Divide-and-Conquer (textittextttSelf-DC) framework, accompanying with the first Compositional unknown Question-Answering dataset (CuQA). This framework enables LLMs to adaptively choose between using internal knowledge and retrieving external knowledge as needed, resulting in a better trade-off between effectiveness and efficiency. Experimental results on two datasets demonstrate that textittextttSelf-DC can achieve comparable or even better performance with much fewer external calls compared with several strong baselines.