# Decoupling Task-Solving and Output Formatting in LLM Generation

arXiv: https://arxiv.org/abs/2510.03595 · HTML: https://mlsanity.com/paper/2510.03595/

## Key facts

- Cited by 9 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 34 of its 45 references that are in the map.
- First posted on arXiv 4 Oct 2025.

## 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 34)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [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)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [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)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Generalization through Memorization: Nearest Neighbor Language Models](https://mlsanity.com/paper/1911.00172/) (2019)
- [Plug and Play Language Models: A Simple Approach to Controlled Text Generation](https://mlsanity.com/paper/1912.02164/) (2019)
- [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://mlsanity.com/paper/1909.05858/) (2019)
- [Cross-Task Generalization via Natural Language Crowdsourcing Instructions](https://mlsanity.com/paper/2104.08773/) (2021)
- [SummEval: Re-evaluating Summarization Evaluation](https://mlsanity.com/paper/2007.12626/) (2020)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [GeDi: Generative Discriminator Guided Sequence Generation](https://mlsanity.com/paper/2009.06367/) (2020)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [Does Prompt Formatting Have Any Impact on LLM Performance?](https://mlsanity.com/paper/2411.10541/) (2024)
- [Token-level Direct Preference Optimization](https://mlsanity.com/paper/2404.11999/) (2024)
- [Guiding LLMs The Right Way: Fast, Non-Invasive Constrained Generation](https://mlsanity.com/paper/2403.06988/) (2024)
- [Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Model](https://mlsanity.com/paper/2310.09520/) (2023)
- [Tractable Control for Autoregressive Language Generation](https://mlsanity.com/paper/2304.07438/) (2023)
- [DIRECTOR: Generator-Classifiers For Supervised Language Modeling](https://mlsanity.com/paper/2206.07694/) (2022)

## Built on by (most cited first, 9 of 9)

- [Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints](https://mlsanity.com/paper/2606.25605/) (2026)
- [The Extrapolation Cliff in On-Policy Distillation of Near-Deterministic Structured Outputs](https://mlsanity.com/paper/2605.08737/) (2026)
- [OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning](https://mlsanity.com/paper/2606.08572/) (2026)
- [One Token Away from Collapse: The Fragility of Instruction-Tuned Helpfulness](https://mlsanity.com/paper/2604.13006/) (2026)
- [It's Not the Capability: Harness Sensitivity Is Non-Monotone Across LLM Agent Tiers](https://mlsanity.com/paper/2605.26731/) (2026)
- [Learning Tractable Distributions Of Language Model Continuations](https://mlsanity.com/paper/2511.16054/) (2025)
- [Arbor: A Framework for Reliable Navigation of Critical Conversation Flows](https://mlsanity.com/paper/2602.14643/) (2026)
- [Decoupled Smart Contract Audits: Lightweight LLM Framework via Distillation and Aggregation](https://mlsanity.com/paper/2606.03128/) (2026)
- [Attributing Structured-Output Gains in Function Calling: Interface Alignment versus Procedural Transfer](https://mlsanity.com/paper/2607.02595/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly adept at solving complex problems, such as mathematical reasoning and automatic evaluation. However, performance often degrades when prompts intertwine task instructions with rigid formatting requirements. This entanglement creates competing goals for the model, hindering its reasoning capabilities. To address this, we introduce Deco-G, a decoding framework that explicitly decouples format adherence from problem solving. Deco-G delegates format adherence to a separate Format Estimation Module (FEM), which performs probabilistic lookahead to estimate future format compliance rate and reweighs token probabilities, allowing the LLM to focus solely on task resolution. To make this approach both practical and efficient, we introduce three key innovations: instruction-aware distillation, a flexible trie-building algorithm, and HMM state pruning. Experiments across mathematical reasoning, event argument extraction, and LLM-as-a-judge demonstrate that Deco-G constantly gains over prompting or structured generation baselines, with guaranteed format compliance. We release our code at https://github.com/haikangdeng/deco-g.