Papers with fine-grained optimization
LongDPO: Unlock Better Long-form Generation Abilities for LLMs via Critique-augmented Stepwise Information (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in large language models have improved their capacity to handle long text inputs, but current models still exhibit unsatisfactory performance in long-form generation. |
| Approach: | They propose a method to enhance long-form text generation through step-level supervision by leveraging Monte Carlo Tree Search to collect stepwise preference pairs and employ a global memory pool to maintain factual accuracy. |
| Outcome: | The proposed method improves performance on long-form generation benchmarks while maintaining lossless performance on several general benchmarks. |
HIPO: A Hierarchical Prompt Optimization Framework with Task Awareness and Fine-Grained Debugging (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for prompt optimization apply the same prompt across all samples . existing methods ignore variation in sample difficulty . |
| Approach: | They propose a framework that shifts the paradigm from dataset-level to sample-level optimization. |
| Outcome: | The proposed framework outperforms baselines on 27 tasks and reduces API calls, token consumption and overall cost by 1.2 to 80. |
ProMed: Shapley Information Gain Guided Reinforcement Learning for Proactive Medical LLMs (2026.acl-long)
Copied to clipboard
Hongxin Ding, Baixiang Huang, Yue Fang, Weibin Liao, Xinke Jiang, Jinyang Zhang, Yinghao Zhu, Zheng Li, Liantao Ma, Junfeng Zhao, Yasha Wang
| Challenge: | Existing medical Large Language Models (LLMs) follow a reactive paradigm, risking diagnostic errors by answering before seeking sufficient details. |
| Approach: | They propose a reinforcement learning framework that transitions LLMs toward a proactive paradigm, enabling them to ask clinically valuable questions before decision-making. |
| Outcome: | Experiments on partial-information medical benchmarks show that ProMed outperforms state-of-the-art methods by 6.29% on average and delivers a 54.45% gain over the reactive paradigm. |