# DIRECT: Direct Decoding for Efficient and Aligned Sequence Labeling with Large Language Models

arXiv: https://arxiv.org/abs/2607.26891 · HTML: https://mlsanity.com/paper/2607.26891/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 16 of its 23 references that are in the map.
- First posted on arXiv 29 Jul 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 16 of 16)

- [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)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [A Survey on Knowledge Graphs: Representation, Acquisition and Applications](https://mlsanity.com/paper/2002.00388/) (2020)
- [Unified Structure Generation for Universal Information Extraction](https://mlsanity.com/paper/2203.12277/) (2022)
- [InstructUIE: Multi-task Instruction Tuning for Unified Information Extraction](https://mlsanity.com/paper/2304.08085/) (2023)
- [GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction](https://mlsanity.com/paper/2310.03668/) (2023)
- [Chinese NER Using Lattice LSTM](https://mlsanity.com/paper/1805.02023/) (2018)
- [Improving Coreference Resolution by Learning Entity-Level Distributed Representations](https://mlsanity.com/paper/1606.01323/) (2016)
- [Universal Information Extraction as Unified Semantic Matching](https://mlsanity.com/paper/2301.03282/) (2023)
- [C-ICL: Contrastive In-context Learning for Information Extraction](https://mlsanity.com/paper/2402.11254/) (2024)
- [Boosting Low-Resource Biomedical QA via Entity-Aware Masking Strategies](https://mlsanity.com/paper/2102.08366/) (2021)
- [Rethinking Negative Instances for Generative Named Entity Recognition](https://mlsanity.com/paper/2402.16602/) (2024)
- [Graphusion: Leveraging Large Language Models for Scientific Knowledge Graph Fusion and Construction in NLP Education](https://mlsanity.com/paper/2407.10794/) (2024)

## Abstract (arXiv)

Sequence labeling is a fine-grained information extraction task, yet existing large language model-based approaches suffer from insufficient domain alignment and low inference efficiency. To address these issues, we propose DIRECT, a framework that addresses these issues through training-time optimization and inference-time rectification. Specifically, DIRECT performs Direct Preference Optimization (DPO) after supervised fine-tuning to strengthen task alignment with human preferences, and introduces a controlled decoding process that enforces fixed output formats and restricts predictions to candidate sets. To further improve efficiency, a template-filling mechanism requires the model to generate only label tokens while reusing prefixed content through the KV Cache, thus reducing redundant computation. Experimental results on eight datasets demonstrate that DIRECT achieves significant improvements in both performance and efficiency compared to existing methods.