Papers with tuning
Continual Learning of Large Language Models (2025.emnlp-tutorials)
Copied to clipboard
| Challenge: | This tutorial explores the challenges of continual learning in large language models . participants will learn strategies to mitigate forgetting and manage data and evaluation pipelines . |
| Approach: | This tutorial offers a comprehensive exploration of continual learning in the context of large language models. |
| Outcome: | This tutorial explores the challenges of continual learning in large language models . participants will learn how to manage data and evaluation pipelines and adapt responsibly . |
UCSMNLP: Statistical Machine Translation for WAT 2019 (D19-52)
Copied to clipboard
| Challenge: | UCSMNLP submitted to WAT 2019 Translation Tasks focusing on Myanmar-English translation. |
| Approach: | They propose to use Name Entity Recognition corpus and bilingual dictionary to build phrase based statistical machine translation system using listwise reranking process and initial distortion weight is changed to improve translation quality. |
| Outcome: | The proposed system outperforms the baseline system in the Myanmar-English translation task. |
Enhancing Visual-Language Modality Alignment in Large Vision Language Models via Self-Improvement (2025.findings-naacl)
Copied to clipboard
Xiyao Wang, Jiuhai Chen, Zhaoyang Wang, Yuhang Zhou, Yiyang Zhou, Huaxiu Yao, Tianyi Zhou, Tom Goldstein, Parminder Bhatia, Taha Kass-Hout, Furong Huang, Cao Xiao
| Challenge: | Existing methods for visual and language alignment depend on external models or data, leading to uncontrollable and unstable results. |
| Approach: | They propose a framework that enhances visual and language alignment without external dependencies by incorporating an in-context self-critic mechanism that constructs preference pairs for tuning. |
| Outcome: | The proposed framework outperforms existing methods and improves performance on 14 hallucination and comprehensive benchmarks. |
BMInf: An Efficient Toolkit for Big Model Inference and Tuning (2022.acl-demo)
Copied to clipboard
Xu Han, Guoyang Zeng, Weilin Zhao, Zhiyuan Liu, Zhengyan Zhang, Jie Zhou, Jun Zhang, Jia Chao, Maosong Sun
| Challenge: | Recent years, pre-trained language models (PLMs) have achieved promising results on various NLP tasks. |
| Approach: | They propose an open-source toolkit for big model inference and tuning which can support big model tuning at extremely low computation cost. |
| Outcome: | The proposed toolkit can support big model inference and tuning at extremely low computation cost. |
MergeIT: From Selection to Merging for Efficient Instruction Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for instruction tuning rely on LLMs to score instruction quality . existing methods rely only on Llms to rank instruction quality, but this approach is expensive and time-consuming . |
| Approach: | They propose a novel LLM-based Merging strategy for better Instruction Tuning that shifts the focus from selection to synthesis. |
| Outcome: | The proposed method reduces time and computational cost while preserving diversity and reducing redundancy. |
SPAGBias: Uncovering and Tracing Structured Spatial Gender Bias in Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are being used in urban planning but there is concern that they reproduce or amplify such biases. |
| Approach: | They propose a framework to evaluate spatial gender bias in large language models . they use a taxonomy of 62 urban micro-spaces, a prompt library and three diagnostic layers . |
| Outcome: | The proposed framework identifies structured gender-space associations that go beyond the public-private divide, forming nuanced micro-level mappings. |
SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe (2026.acl-long)
Copied to clipboard
| Challenge: | Efforts to improve instruction tuning often focus on higher-quality supervised fine-tuning datasets, typically requiring data filtering with proprietary LLMs or human annotation. |
| Approach: | They propose a Mixup-based recipe that elevates LLM instruction tuning without relying on well-curated datasets. |
| Outcome: | The proposed model improves instruction-following and healthcare-specific tasks with consistent improvements across LLM families and SFT datasets. |
Dynamic Data Mixing Maximizes Instruction Tuning for Mixture-of-Experts (2025.naacl-long)
Copied to clipboard
| Challenge: | Mixture-of-Experts (MoE) models are constrained by their fixed model capacities when the number of tasks grows in instruction tuning. |
| Approach: | They propose to combine all training tasks and apply fixed sampling weights without considering the importance of different tasks as the model training state changes. |
| Outcome: | The proposed method can be used on knowledge & reasoning tasks and open-ended queries with limited training budget. |
I-SEE: An Instruction-tuned, SOP-Enhanced Quality Evaluator for Product Content (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing approaches to content evaluation treat information uniformly without prioritizing based on customer relevance. |
| Approach: | They propose a framework that combines domain expertise with a single instruction to improve content. |
| Outcome: | a new framework outperforms existing models in detecting inconsistencies across 20 product categories and 150 product specific features. |
Sorted LLaMA: Unlocking the Potential of Intermediate Layers of Large Language Models for Dynamic Inference (2024.findings-eacl)
Copied to clipboard
| Challenge: | Large language models excel at understanding and generating human-like text, but their widespread deployment can be prohibitively expensive. |
| Approach: | They propose a method that makes large language models dynamic without Pre-Training . they use modularity in networks and sort sub-models based on computation/accuracy in a nested manner. |
| Outcome: | The proposed method can make large language models dynamic without pre-training and replace standard fine-tuning with sorted fine- tuning. |
LSDC: An Efficient and Effective Large-Scale Data Compression Method for Supervised Fine-tuning of Large Language Models (2025.findings-naacl)
Copied to clipboard
Zhaoguang Long, Yuhao Zhou, Shangqing Zhao, Yupei Ren, Li Cai, Chenghao Jia, Zhe Chen, Zhe Fang, Yuxiang Song, Man Lan
| Challenge: | Large Language Models (LLMs) are expanding in scale and size, increasing computational costs . large-scale data compression techniques can reduce the size of training datasets while maintaining data integrity. |
| Approach: | They propose a large-scale data compression method to reduce the size of training data . they use a bifurcated quantization strategy to maximize the diversity of samples . |
| Outcome: | The proposed method significantly reduces the size of training data while maximizing the submodular gain. |
LLM Sensitivity Challenges in Abusive Language Detection: Instruction-Tuned vs. Human Feedback (2025.coling-main)
Copied to clipboard
| Challenge: | Existing studies show that instruction-tuned LLMs under-predict positive classes . however, they are overly sensitive and can be applied for abuse detection without fine-tuning . |
| Approach: | They show that instruction-tuned LLMs tend to under-predict positive classes . they also show that label frequency in the prompt helps with the significant over-prediction . |
| Outcome: | The proposed models under-predict positive classes in social media, whereas they are overly sensitive. |
Data Diversity Matters for Robust Instruction Tuning (2024.findings-emnlp)
Copied to clipboard
Alexander Bukharin, Shiyang Li, Zhengyang Wang, Jingfeng Yang, Bing Yin, Xian Li, Chao Zhang, Tuo Zhao, Haoming Jiang
| Challenge: | Recent studies have shown that by curating high quality and diverse instruction tuning datasets, we can significantly improve instruction-following capabilities. |
| Approach: | They propose an algorithm to control diversity and quality of instruction tuning datasets and validate it. |
| Outcome: | The proposed algorithm significantly improves worst and average case performance on large scale instruction tuning datasets. |
Stronger Models are Not Always Stronger Teachers for Instruction Tuning (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to optimize instruction-following capabilities of large language models (LLMs) assume that larger or stronger models are stronger teachers and therefore adopt smaller models as response generators. |
| Approach: | They propose to use large-scale instruction datasets to tune large language models to align with specific tasks and user intents. |
| Outcome: | The proposed metric outperforms most baselines in identifying the effectiveness of response generators. |
Respond in my Language: Mitigating Language Inconsistency in Response Generation based on Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) show strong instruction understanding ability across multiple languages, but are easily biased towards English in instruction tuning. |
| Approach: | They propose to use a model with Pseudo-Inconsistent Penalization to prevent the model from generating English responses when given non-English language prompts during training and prior Enhanced decoding to improve the language consistency of the model. |
| Outcome: | The proposed methods significantly improve the language consistency of the model without multilingual data. |
Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for instruction tuning do not include associating instructions with existing datasets. |
| Approach: | They propose a dynamic growth paradigm for the automatic curation of instruction-tuning data . they use existing datasets to automatically construct instruction-uning datasets . |
| Outcome: | The proposed model reduces the API cost for generating instructions and provides high-quality data. |
Using J-K-fold Cross Validation To Reduce Variance When Tuning NLP Models (C18-1)
Copied to clipboard
| Challenge: | a recent study shows that performance estimations are unstable and variable . this makes it difficult to use parameter tuning and model selection . |
| Approach: | They propose to use a less variable CV method to evaluate performance . they propose lower choices of K than are typically seen in the NLP literature . |
| Outcome: | The proposed method can be used for parameter tuning and performance estimation, but it is unstable and unstable. |
One-Shot Learning as Instruction Data Prospector for Large Language Models (2024.acl-long)
Copied to clipboard
Yunshui Li, Binyuan Hui, Xiaobo Xia, Jiaxi Yang, Min Yang, Lei Zhang, Shuzheng Si, Ling-Hao Chen, Junhao Liu, Tongliang Liu, Fei Huang, Yongbin Li
| Challenge: | Contemporary practices in instruction tuning often hinge on enlarging data scaling without a clear strategy for ensuring data quality. |
| Approach: | They propose a method that leverages one-shot learning to discern and select high-quality instruction data from extensive datasets. |
| Outcome: | Nuggets outperforms existing methods on MT-Bench and Alpaca-Eval benchmarks. |
Multi-Objective Linguistic Control of Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing Large language models prefer to generate verbose responses due to the length bias, which may increase unnecessary reading complexity. |
| Approach: | They propose to use off-the-shelf data to fine tune multiple linguistic complexities of LLM outputs to improve multi-complexity controllability and improve the quality of the responses. |
| Outcome: | The proposed method improves multi-complexity controllability significantly and retains or enhances the quality of the responses as a side benefit. |
Chat-TS: Enhancing Multi-Modal Reasoning Over Time-Series and Natural Language Data (2026.eacl-long)
Copied to clipboard
| Challenge: | Large language models are being rapidly applied across many fields such as healthcare, finance, transportation, and energy. |
| Approach: | They propose a large language model framework that integrates time-series tokens into LLMs’ vocabulary, enhancing its reasoning ability over time- and textual data. |
| Outcome: | The proposed framework enhances reasoning ability over time-series and textual data without compromising core natural language capabilities. |
Scalable Fine-tuning from Multiple Data Sources: A First-Order Approximation Approach (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a new algorithm to estimate fine-tuning performance for a target task is proposed . conventional subset selection methods require repeated training on subsets of auxiliary tasks . |
| Approach: | They propose an algorithm to fine-tune a language model for a target task by optimally using auxiliary tasks' information. |
| Outcome: | The proposed method can estimate fine-tuning performance on CPUs in seconds. |
TableLlama: Towards Open Large Generalist Models for Tables (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for interpreting, augmenting, and querying semi-structured tables require pretraining on tables or special model architecture design. |
| Approach: | They construct a dataset with a variety of tables and tasks for instruction tuning and evaluating LLMs. |
| Outcome: | The proposed model achieves comparable or better performance on 7 out of 8 in-domain tasks compared with the base model on 6 out-of-domain datasets. |
From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora (2025.emnlp-main)
Copied to clipboard
| Challenge: | Experiments show that models trained on multi-way parallel data outperform those trained on unaligned data. |
| Approach: | They propose a large-scale, high-quality multi-way parallel corpus based on TED Talks that spans 113 languages with up to 50 languages aligned in parallel. |
| Outcome: | The proposed model outperforms models trained on unaligned multilingual data on six multilingual benchmarks. |
Are Large Language Model Temporally Grounded? (2024.naacl-long)
Copied to clipboard
| Challenge: | Recent large language models lack a consistent temporal model of textual narratives . sentence ordering in unlabelled texts is only weakly correlated with event ordering . |
| Approach: | They evaluate LLMs with textual narratives and evaluate their common-sense knowledge . they find that LLM models struggle the most with self-consistency . |
| Outcome: | The proposed models lack a consistent temporal model of textual narratives. |
Grammar Induction with Neural Language Models: An Unusual Replication (D18-1)
Copied to clipboard
| Challenge: | Recent work on latent tree learning attempts to develop models with parse-valued latent variables and train them on non-parsing tasks. |
| Approach: | They propose a model with parse-valued latent variables and a strong latent tree learning result on constituency parsing. |
| Outcome: | The proposed model outperforms all baselines and performs competitively with symbolic grammar induction systems. |
Improving domain-specific SMT for low-resourced languages using data from different domains (L18-1)
Copied to clipboard
| Challenge: | Evaluation of domain-specific statistical machine translation system for official government letters . use of pseudo in-domain data showed improvement for both test sets . |
| Approach: | They develop a statistical machine translation system for official government letters . the system is based on a parallel in-domain dataset containing official letters based in Sinhala and Tamil . |
| Outcome: | The proposed system improves on the in-domain data in the domain of official government letters . the evaluations show that the system requires quality data from diverse subject matters and sources to perform better. |
Teaching According to Talents! Instruction Tuning LLMs with Competence-Aware Curriculum Learning (2025.findings-emnlp)
Copied to clipboard
Yangning Li, Tingwei Lu, Yinghui Li, Yankai Chen, Wei-Chieh Huang, Wenhao Jiang, Hui Wang, Hai-Tao Zheng, Philip S. Yu
| Challenge: | Efficient instruction tuning aims to enhance the ultimate performance of large language models (LLMs) current methods suffer from the curriculum rigidity, resulting in a fixed and potentially sub-optimal learning trajectory. |
| Approach: | a framework for efficient instruction tuning is proposed to address the issue of curriculum rigidity . current methods rely on static heuristic difficulty metrics and fail to adapt to evolving capabilities . |
| Outcome: | Efficient instruction tuning aims to enhance the ultimate performance of large language models . current methods suffer from the curriculum rigidity, resulting in a fixed learning trajectory . |
Pipeline Analysis for Developing Instruct LLMs in Low-Resource Languages: A Case Study on Basque (2025.naacl-long)
Copied to clipboard
| Challenge: | Large language models are typically optimized for resource-rich languages like English . however, the proprietary nature of these models makes them impractical for many researchers and developers. |
| Approach: | They propose to develop large language models that can follow instructions in Basque . they focus on three key stages: pre-training, instruction tuning, and alignment with human preferences . |
| Outcome: | The proposed models improve natural language understanding (NLU) of the foundational model by 12 points . the results show that the models can follow instructions in Basque with human preferences . |
A Survey on MLLM-based Visually Rich Document Understanding: Methods, Challenges, and Emerging Trends (2026.findings-acl)
Copied to clipboard
Yihao Ding, Siwen Luo, Yue Dai, Yanbei Jiang, Zechuan Li, Qiang Sun, Geoffrey Martin, Wei Liu, Yifan Peng
| Challenge: | Visually Rich Document Understanding (VRDU) frameworks are a key area of research . early approaches to VRDU relied on manually crafted rules and domain-specific heuristics . conventional deep learning approaches do not integrate the diverse modalities in documents . |
| Approach: | They review recent advances in MLLM-based Visually Rich Document Understanding (VRDU) their findings highlight emerging trends and promising research directions . |
| Outcome: | The proposed frameworks are scalable, reliable, and adaptable, the authors argue . their findings highlight emerging trends and promising research directions . |
Differentiable Instruction Optimization for Cross-Task Generalization (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have shown that instruction tuning is effective for generalizing to arbitrary tasks unseen during training. |
| Approach: | They propose to introduce learnable instructions and optimize them with gradient descent to optimize instruction for generalization ability. |
| Outcome: | The proposed instruction extractor extracts appropriate instruction and improves generalization ability compared to manual instruction tuning. |
Bridging the Memorization-Utilization Gap: Near-Lossless Context Compression via Reinforcement Learning (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in context compression have failed to effectively utilize compressed representations for downstream tasks. |
| Approach: | They propose a holistic training paradigm that uses outcome-based RL to enable implicit expansion. |
| Outcome: | The proposed model outperforms previous models on NIAH, LongBench and multi-hop reasoning. |
InstructSafety: A Unified Framework for Building Multidimensional and Explainable Safety Detector through Instruction Tuning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing safety detection systems have limitations in terms of their versatility and interpretability. |
| Approach: | They introduce a safety detection framework that unifies 7 common sub-tasks into a uniform formulation and process 39 human-annotated datasets for instruction tuning. |
| Outcome: | The proposed framework unifies 7 common sub-tasks into a uniform formulation and then runs on 39 human-annotated datasets to fine-tune it. |
Learning to Generate Instruction Tuning Datasets for Zero-Shot Task Adaptation (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing instruction tuning datasets are limited by the quality of the instruction tuning data. |
| Approach: | They propose a model that converts unannotated text into task-specific training datasets for instruction tuning. |
| Outcome: | The proposed model improves the performance of pretrained and instruction tuned models over the de facto self-supervised baseline. |
Superfiltering: Weak-to-Strong Data Filtering for Fast Instruction-Tuning (2024.acl-long)
Copied to clipboard
| Challenge: | Earlier studies of instruction tuning on Large Language Models focus on creating large, varied, and high-quality datasets with responses curated by human experts. |
| Approach: | They propose to use a smaller and weaker model to fine tune a larger and stronger model . they find it can largely speed up the data filtering and improve performance . |
| Outcome: | The proposed model can filter instruction data faster and better on benchmarks. |
NativQA: Multilingual Culturally-Aligned Natural Query for LLMs (2025.findings-acl)
Copied to clipboard
Md. Arid Hasan, Maram Hasanain, Fatema Ahmad, Sahinur Rahman Laskar, Sunaya Upadhyay, Vrunda N Sukhadia, Mucahid Kutlu, Shammur Absar Chowdhury, Firoj Alam
| Challenge: | Existing frameworks for QA datasets lack regional specificity and cultural specificity. |
| Approach: | They propose a framework to quench native language QA datasets in native languages for LLM evaluation and tuning. |
| Outcome: | The proposed framework is scalable, language-independent and can be used to build culturally and regionally aligned QA datasets in native languages. |
InstructGraph: Boosting Large Language Models via Graph-centric Instruction Tuning and Preference Alignment (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing large language models (LLMs) can solve graph reasoning and generation tasks with parameter updates without sacrificing performance. |
| Approach: | They propose a structured format verbalizer to unify all graph data into a universal code-like format, which can simply represent the graph without any external graph-specific encoders. |
| Outcome: | The proposed framework outperforms GPT-4 and LLaMA2 in graph reasoning and generation tasks by more than 13% and 38%, respectively. |
An Empirical Analysis of Parameter-Efficient Methods for Debiasing Pre-Trained Language Models (2023.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models inherit more human-like biases from the training corpora, causing computationally expensive problems. |
| Approach: | They propose parameter-efficient methods in combination with counterfactual data augmentation for bias mitigation. |
| Outcome: | The proposed methods are effective in mitigating gender bias, prompt tuning is more suitable for GPT-2 than BERT, and less effective when it comes to racial and religious bias. |
Deep Exploration of Cross-Lingual Zero-Shot Generalization in Instruction Tuning (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have focused on instruction tuning to show cross-lingual generalization . a novel non-English meta-dataset is used to study instruction tuning . |
| Approach: | They perform instruction tuning individually for two distinct language meta-datasets and assess the performance on unseen tasks in a non-English language. |
| Outcome: | The proposed model outperforms baseline training in English and Korean by 20.7% and 13.6%. |
GAP: a Global Adaptive Pruning Method for Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing structured pruning methods employ uniform compression rates across network layers, neglecting the varying importance of different network depths. |
| Approach: | They propose a pruning framework that minimizes global capability loss by layer-adaptive pruning rates. |
| Outcome: | The proposed approach achieves comparable performance with state-of-the-art methods at high pruning rates and shows significant advantages at low pruning rates. |
Improving Preference Alignment of LLM with Inference-Free Self-Refinement (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) develop in-context learning capability through pretraining and instruction tuning. |
| Approach: | Large language models (LLMs) develop in-context learning capability through pretraining and instruction tuning. |
| Outcome: | Experiments show that incorporating IFSR into preference alignment yields performance improvement over 10%. |
Probing and Boosting Large Language Models Capabilities via Attention Heads (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to identifying capabilities rely on external signals with limited structural grounding . emergence of specific capabilities remains poorly understood . |
| Approach: | They propose a lightweight approach that links LLM capabilities to internal components by identifying correspondences at the level of attention heads. |
| Outcome: | The proposed approach improves accuracy on MMLU and BBH by 1 to 1.5 points over gradient-based method and 5 to 6 points over other intermediate-state baselines. |
SSSD: Simply-Scalable Speculative Decoding (2026.acl-long)
Copied to clipboard
Michele Marzollo, Jiawei Zhuang, Niklas Roemer, Niklas Zwingenberger, Lorenz K Muller, Lukas Cavigelli
| Challenge: | Existing methods for accelerating inference in Large Language Models require additional training and training, resulting in a higher deployment and maintenance cost. |
| Approach: | They propose a training-free method that combines lightweight n-gram matching with hardware-aware speculation. |
| Outcome: | SSSD reduces latency by up to 2.9 and is faster than autoregressive decoding methods. |
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. |