Papers with optimization methods
CoLLiE: Collaborative Training of Large Language Models in an Efficient Way (2023.emnlp-demo)
Copied to clipboard
Kai Lv, Shuo Zhang, Tianle Gu, Shuhao Xing, Jiawei Hong, Keyu Chen, Xiaoran Liu, Yuqing Yang, Honglin Guo, Tengxiao Liu, Yu Sun, Qipeng Guo, Hang Yan, Xipeng Qiu
| Challenge: | Large language models (LLMs) are increasingly pivotal in a wide range of tasks . however, the resources required for training these models necessitate efficient solutions . |
| Approach: | They propose a library that facilitates collaborative training of large language models . they use 3D parallelism, parameter-efficient fine-tuning methods and optimizers . |
| Outcome: | The proposed library has proven superior training efficiency in comparison with prevalent solutions in pre-training and fine-tuning scenarios. |
Global Optimization under Length Constraint for Neural Text Summarization (P19-1)
Copied to clipboard
| Challenge: | GOLC increases the probabilities of generating summaries that have high evaluation scores within a desired length. |
| Approach: | They propose a global optimization method under length constraint for neural text summarization models. |
| Outcome: | The proposed method generates fewer overlength summaries while maintaining the fastest processing speed. |
Rethinking Action Spaces for Reinforcement Learning in End-to-end Dialog Agents with Latent Variable Models (N19-1)
Copied to clipboard
| Challenge: | Existing approaches to define action spaces for conversational agents have limitations . end-to-end dialog systems can handle complex domains with limited action space . |
| Approach: | They propose a latent action framework that treats the action spaces of an end-to-end dialog agent as latent variables and develops unsupervised methods to induce its own action space from the data. |
| Outcome: | The proposed framework achieves better performance than word-level policy gradient methods on DealOrNoDeal and MultiWoz dialogs. |
Safely Learning with Private Data: A Federated Learning Framework for Large Language Model (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing large language models (LLMs) use large amounts of public data and massive parameters, but private data is often stored in isolated data silos. |
| Approach: | They propose a Federated Learning framework for large language models which offloads most training parameters to the server while training embedding and output layers locally. |
| Outcome: | The proposed framework achieves comparable metrics to centralized chatGLM model on NLU and generation tasks. |
Optimized Speculative Sampling for GPU Hardware Accelerators (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large foundational speech and language models require more memory and computational resources to generate long sequences. |
| Approach: | They propose to optimize speculative sampling for parallel hardware accelerators by combining multiple GPU threads to reduce profiling time. |
| Outcome: | The proposed approach improves profiling time from 6% to 13% without compromising accuracy. |
A Practical Analysis of Human Alignment with *PO (2025.findings-naacl)
Copied to clipboard
| Challenge: | Prior research focused on identifying the best-performing method to varying hyperparameters . prior research focused primarily on a grid search, which can be impractical for general practitioners . |
| Approach: | They propose a preference optimization method that is more stable across hyperparameters and reduces the average response length. |
| Outcome: | The proposed method increases likelihood of achieving better results through various metrics, such as KL divergence and response length. |
Towards Optimal Evaluation Efficiency for Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) require large-scale benchmarks, which are costly in terms of time, computational resources, or API tokens. |
| Approach: | They propose an efficient evaluation framework that selects a question subset based on pre-tested results and uses semantic analysis to evaluate whether the subset preserves the original benchmark. |
| Outcome: | The proposed evaluation framework outperforms previous methods in reliability and score accuracy. |
SCULPT: Systematic Tuning of Long Prompts (2025.acl-long)
Copied to clipboard
Shanu Kumar, Akhila Yesantarao Venkata, Shubhanshu Khandelwal, Bishal Santra, Parag Agrawal, Manish Gupta
| Challenge: | Existing methods for prompt optimization struggle with longer, more complex ones, often risking information loss and being sensitive to small perturbations. |
| Approach: | They propose a framework that treats prompt optimization as a hierarchical tree refinement problem and uses a Critic-Actor framework to generate reflections and apply actions to refine the prompt. |
| Outcome: | The proposed framework produces more stable and interpretable prompt modifications, ensuring better generalization across tasks. |
Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation (2026.acl-long)
Copied to clipboard
| Challenge: | Debiased large language models excel at handling known or low-bias prompts, but fail on unfamiliar and high-biased prompts. |
| Approach: | They propose a debiasing framework that detects high-bias prompts and triggers context-aware LoRA updates only when a bias-risk score exceeds a threshold. |
| Outcome: | The proposed framework reduces toxicity/bias score with significantly lower latency than standard optimization methods. |
Think Before Writing: Feature-Level Multi-Objective Optimization for Generative Citation Visibility (2026.acl-long)
Copied to clipboard
| Challenge: | Existing generative engine optimization approaches rely on token-level text rewriting, offering limited interpretability and weak control over the trade-off between visibility and content quality. |
| Approach: | They propose a feature-level, multi-objective optimization framework that abstracts webpages into interpretable structural, content, and linguistic properties. |
| Outcome: | The proposed framework outperforms token-level methods in citation visibility and content quality on three generative engines. |
Adaptive Spatial and Temporal Redundancy Optimization for Efficient Reasoning in Large Language Models (2026.acl-long)
Copied to clipboard
Tianle Chen, Pengyu Cheng, Qiyuan Zhu, Jiacheng Wang, Bei Liu, Hao Gu, Ruijie Shen, Xiaofeng Hou, Sirui Han, Jiacheng Liu
| Challenge: | Existing research to improve CoT efficiency falls into three categories, each with distinct limitations. |
| Approach: | They propose a training-free framework that addresses both dimensions of CoT reasoning by applying a progressive precision reduction strategy coupled with an entropy-based confidence mechanism for adaptive termination. |
| Outcome: | Empirical results show that the proposed framework achieves 11.3 efficiency gain without compromising accuracy. |
Compound AI Systems Optimization: A Survey of Methods, Challenges, and Future Directions (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) and AI systems have led to a paradigm shift in the design and optimization of complex workflows. |
| Approach: | They propose a systematic review of recent progress in optimizing compound AI systems . they formalize the notion of compound AI system optimization and classify existing methods along several key dimensions . |
| Outcome: | The proposed methods outperform existing methods in the field of compound AI and highlight open research challenges and future directions. |
PlanE: Meta Planning of Data, Tuning, and Inference for Extractive-based LLMs (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for optimizing LLMs for task-specific tasks are limited due to the sheer volume of data. |
| Approach: | They propose a Planning framework for constructing Extractive-based LLMs called PlanE . they propose 'data decomposition', instruction tuning, prompt inference and a 'Data-Tuning-Inference' planner . |
| Outcome: | The proposed framework improves performance across different datasets and on different dataset. |