Papers with data curation
Gradient-Attention Guided Dual-Masking Synergetic Framework for Robust Text-based Person Retrieval (2025.emnlp-main)
Copied to clipboard
| Challenge: | a large-scale visionlanguage pre-training framework is limited by the scarcity of large-sized annotated vision-language data . noise-resistant data construction pipeline is needed to filter and caption web-sourced images . noisy text tokens can be a problem for fine-grained representation learning . |
| Approach: | They develop a noise-resistant data construction pipeline that leverages in-context learning capabilities of MLLMs to automatically filter and caption web-sourced images. |
| Outcome: | The proposed framework improves cross-modal alignment by masking noisy textual tokens based on the gradient-attention similarity score. |
Developing Japanese CLIP Models Leveraging an Open-weight LLM for Large-scale Dataset Translation (2025.naacl-srw)
Copied to clipboard
| Challenge: | lack of large-scale open Japanese image-text pairs poses a significant barrier to the development of vision-language models. |
| Approach: | They construct large-scale Japanese image-text pairs using machine translation and pre-trained CLIP models on a Japanese dataset. |
| Outcome: | The results show that pre-trained models achieve competitive average scores on Japanese culture tasks compared to models of similar size. |
DialogGen: Multi-modal Interactive Dialogue System with Multi-turn Text-Image Generation (2025.findings-naacl)
Copied to clipboard
Minbin Huang, Yanxin Long, Xinchi Deng, Ruihang Chu, Jiangfeng Xiong, Xiaodan Liang, Hong Cheng, Qinglin Lu, Wei Liu
| Challenge: | Text-to-image (T2I) generation models have advanced in recent years, but effective interaction with these models is challenging for average users due to the need for specialized prompt engineering knowledge and the inability to perform multi-turn image generation. |
| Approach: | They propose to use off-the-shelf MLLMs and T2I models to build a multi-modal interactive dialogue system (MIDS) that can generate correct output modalities and coherence of output images. |
| Outcome: | The proposed pipeline can generate correct output modalities and coherent multi-modal outputs compared with other state-of-the-art models. |
That doesn’t sound right: Evaluating speech transcription quality in field linguistics corpora (2025.acl-short)
Copied to clipboard
| Challenge: | Automated speech recognition (ASR) is a popular tool for documenting languages, but field linguists do not have the data to train robust models. |
| Approach: | They propose to use fieldwork data to identify speech transcriptions that may be unsuitable for training ASR models. |
| Outcome: | The proposed measures can be used to identify transcriptions with characteristics common in field data but could be detrimental to ASR training. |
Bhaasha, Bhāṣā, Zaban: A Survey for Low-Resourced Languages in South Asia – Current Stage and Challenges (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a survey examines the current efforts and challenges of NLP models for South Asian languages . there are more than 650 languages in South Asia, but many have very limited computational resources or are missing from existing models. |
| Approach: | a survey examines efforts and challenges of NLP for South Asian languages . they focus on transformer-based models such as BERT, T5, & GPT . findings highlight substantial issues, including missing data in critical domains . |
| Outcome: | The findings highlight significant issues, including missing data in critical domains . the survey aims to raise awareness within the NLP community for more targeted data curation . |
FinRAG-12B: A Production-Validated Recipe for Grounded Question Answering in Banking (2026.acl-industry)
Copied to clipboard
| Challenge: | Large language models (LLMs) are rapidly being adopted across various domains, but adoption in the regulated banking industry is limited due to their tendency to hallucinate, exhibit over-agreeable behavior, and lack alignment with domain-specific knowledge and constraints. |
| Approach: | They propose a framework for training grounded domain-specific LLMs that optimizes answer quality, citation grounding, and calibrated refusal under real-world deployment constraints. |
| Outcome: | The proposed model outperforms GPT-4.1 on citation grounding and calibrated refusal under real-world deployment constraints. |
AugESC: Dialogue Augmentation with Large Language Models for Emotional Support Conversation (2023.findings-acl)
Copied to clipboard
| Challenge: | Crowdsourced dialogue corpora are limited in scale and topic coverage due to the expensive cost of data curation. |
| Approach: | They construct an augmented dataset for the emotional support conversation task using large language models for dialogue augmentation. |
| Outcome: | The proposed approach outperforms baselines of dialogue augmentation and improves the model's generalization ability to open-domain topics. |
GlobalWoZ: Globalizing MultiWoZ to Develop Multilingual Task-Oriented Dialogue Systems (2022.acl-long)
Copied to clipboard
| Challenge: | Existing multilingual task-oriented dialogue datasets lack high-quality data curation due to the high expense and challenges of human annotation. |
| Approach: | They propose a method that generates a multilingual ToD dataset globalized from an English ToD data set for three unexplored use cases of multilingual toD systems. |
| Outcome: | The proposed method generates a large-scale multilingual ToD dataset globalized from an English ToD data set for three unexplored use cases of multilingual toD systems. |
ABCD-LINK: Annotation Bootstrapping for Cross-Document Fine-Grained Links (2026.eacl-long)
Copied to clipboard
| Challenge: | Using retrieval models and LLMs achieves a 73% approval rate for suggested links, more than doubling the acceptance of strong retrievers alone. |
| Approach: | They propose a domain-agnostic framework for bootstrapping sentence-level cross-document links from scratch and apply it to large-scale human-in-the-loop annotation of natural text pairs. |
| Outcome: | The proposed framework generates semi-synthetic datasets and uses them to benchmark and shortlist the best-performing methods and applies them in large-scale human-in-the-loop annotation of natural text pairs. |
Changing the World by Changing the Data (2021.acl-long)
Copied to clipboard
| Challenge: | a new paper argues that data curation is already happening, and it is changing the world . social biases and spurious patterns are attracting more attention in NLP models . |
| Approach: | They argue that data curation is already happening and will be happening . they argue that social biases and spurious patterns are the main problems . |
| Outcome: | a new paper argues that data curation is already and will be happening, and it is changing the world. |
Bringing Real-World Relations into Video Generation with Graph-Structured Knowledge (2026.acl-long)
Copied to clipboard
| Challenge: | Existing text-to-video models struggle to accurately simulate real-world physics and dynamic entity interactions. |
| Approach: | They propose a framework that integrates graph-structured temporal knowledge into video latent diffusion models to enhance compositional generation and interaction fidelity. |
| Outcome: | The proposed framework enhances compositional generation and interaction fidelity by integrating graph-structured temporal knowledge into video latent diffusion models. |
Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction (2024.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models may not follow human instructions and produce toxic, hallucinated, or biased content. |
| Approach: | They propose a disperse-then-merge framework that dispersers instruction-following data into portions and trains multiple sub-models using different data portions. |
| Outcome: | The proposed framework outperforms data curation and training regularization on standard knowledge and reasoning benchmarks. |
Nanda Family: Open-Weights Generative Large Language Models for Hindi (2026.eacl-long)
Copied to clipboard
Aaryamonvikram Singh, Debopriyo Banerjee, Dhruv Sahnan, Monojit Choudhury, Shivam Chauhan, Rocktim Jyoti Das, Xudong Han, Haonan Li, Alok Anil Jadhav, Utkarsh Agarwal, Mukund Choudhary, Fajri Koto, Junaid Hamid Bhat, Awantika Shukla, Samujjwal Ghosh, Samta Kamboj, Onkar Pandit, Lalit Pradhan, Rahul Pal, Sunil Kumar Sahu, Parvez Mullah, Ali El Filali, Zainul Abedien Ahmed Quraishi, Neha Sengupta, Gokulakrishnan Ramakrishnan, Rituraj Joshi, Gurpreet Gosal, Avraham Sheinin, Natalia Vassilieva, Preslav Nakov
| Challenge: | Large language models remain predominantly English-centric, which limits their utility for underrepresented languages. |
| Approach: | They propose to extend Llama’s vocabulary with 20% Hindi-specific tokens, thus halving Hindi tokenization fertility while preserving English efficiency. |
| Outcome: | The proposed models outperform open-weight models of comparable size on a 65B-token corpus and bilingual instruction and safety alignment on . a culturally grounded dataset. |
Revisiting Generalization Across Difficulty Levels: It’s Not So Easy (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing research is mixed regarding whether training on easier or harder data leads to better results. |
| Approach: | They examine how well large language models generalize across different task difficulties by using a large dataset and a well-established difficulty metric. |
| Outcome: | The results show that training on hard data can't achieve consistent improvements across the full range of difficulties. |
LLM2LLM: Boosting LLMs with Novel Iterative Data Enhancement (2024.findings-acl)
Copied to clipboard
Nicholas Lee, Thanakul Wattanawong, Sehoon Kim, Karttikeya Mangalam, Sheng Shen, Gopala Anumanchipalli, Michael Mahoney, Kurt Keutzer, Amir Gholami
| Challenge: | Pretrained large language models are currently state-of-the-art for solving most tasks . however, many of them are in the low-data regime, making fine-tuning challenging . a new data augmentation strategy uses a teacher LLM to augment a small seed dataset . |
| Approach: | They propose a targeted and iterative data augmentation strategy that augments a teacher LLM to fine-tune a small seed dataset by adding additional data. |
| Outcome: | The proposed approach outperforms fine-tuning and other data augmentation strategies on a small seed dataset. |
AboutMe: Using Self-Descriptions in Webpages to Document the Effects of English Pretraining Data Filters (2024.acl-long)
Copied to clipboard
| Challenge: | Large language models' (LLMs) abilities are drawn from their pretraining data. however, decisions around what data is retained or removed during this initial stage are under-scrutinized. |
| Approach: | They ground web text, a popular pretraining data source, to its social and geographic contexts. |
| Outcome: | The results show that some quality classifiers act like topical domain filters, and langID overlook English content from some regions of the world. |
Interoperability in an Infrastructure Enabling Multidisciplinary Research: The case of CLARIN (2020.lrec-1)
Copied to clipboard
| Challenge: | CLARIN supports the use and study of language data in general and aims to increase the potential for comparative research of cultural and societal phenomena across languages and disciplines. |
| Approach: | They describe the interoperability requirements that arise through the existing ambitions and emerging frameworks. |
| Outcome: | The proposed frameworks will address interoperability requirements at several levels, including organisation and ecosystem, design of workflow services, data curation, performance measurement and collaboration. |
Learning to Describe Implicit Changes: Noise-robust Pre-training for Image Difference Captioning (2025.findings-emnlp)
Copied to clipboard
Zixin Guo, Jiayang Sun, Tzu-Jui Julius Wang, Abduljalil Radman, Selen Pehlivan, Min Cao, Jorma Laaksonen
| Challenge: | Large Multimodal Models (LMMs) are used to capture subtle differences between images but are noisy and coarse summaries. |
| Approach: | They propose a noise-robust approach to image difference capture using large multimodal models . they use LMMs with structured prompts to generate fine-grained change descriptions . |
| Outcome: | The proposed model outperforms streamlined architectures and improves inference efficiency. |
Reconstructing NER Corpora: a Case Study on Bulgarian (2020.lrec-1)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) and Named Enel Linking (NEL) are two related tasks that are under-resourced for the Slavic languages. |
| Approach: | They propose to use deep learning methods to improve a Named Entity Recognition corpus and to predict and annotate new types in a test corpus. |
| Outcome: | The proposed model improves a type-based Named Entity Recognition (NER) training corpus and predicts and annotates new types in a test corpus. |
Position Paper: Data-Centric AI in the Age of Large Language Models (2024.findings-emnlp)
Copied to clipboard
Xinyi Xu, Zhaoxuan Wu, Rui Qiao, Arun Verma, Yao Shu, Jingtan Wang, Xinyuan Niu, Zhenfeng He, Jiangwei Chen, Zijian Zhou, Gregory Kang Ruey Lau, Hieu Dao, Lucas Agussurja, Rachael Hwee Ling Sim, Xiaoqiang Lin, Wenyang Hu, Zhongxiang Dai, Pang Wei Koh, Bryan Kian Hsiang Low
| Challenge: | a paper proposes a data-centric perspective of AI research, focusing on large language models. |
| Approach: | They propose a data-centric viewpoint of AI research, focusing on large language models . they propose four scenarios centered around data, including data curation, attribution, knowledge transfer . |
| Outcome: | The proposed research focuses on large language models with data centric benchmarks . the proposed benchmarks can be used to develop new data curation methods . |
LexMatcher: Dictionary-centric Data Curation for LLM-based Machine Translation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | emergence of large language models (LLMs) has brought about new opportunities for machine translation. |
| Approach: | They propose a method for data curation that supplements the infrequent senses of polysemous words. |
| Outcome: | The proposed method outperforms established baselines on the WMT2022 test sets and is applicable to other pre-trained models. |
RST-Guarder: Enhancing Long-Context Robustness for Safeguards via RST Parsing and Probabilistic Inference (2026.acl-long)
Copied to clipboard
| Challenge: | Existing guardrail models for harmful-content detection degrade on long-form inputs . Existing models are vulnerable to policy-violating responses, causing false positives based on benign content . |
| Approach: | They propose an inference-time method that improves harmful-content detection for long-form inputs without additional data curation or model training. |
| Outcome: | The proposed method improves harmful-content detection for long-form inputs without additional data curation or model training. |
UniGeM: Unifying Data Selection and Mixing via Geometric Exploration and Mining (2026.findings-acl)
Copied to clipboard
Changhao Wang, null Yunfeiyu, Xinhao Yao, Jiaolong Yang, Lu Yu, Junpeng Fang, Chaobo Li, Riccardo Cantoro, Qing Cui, Jun Zhou
| Challenge: | Large Language Models (LLMs) scaling is limited by data quality and domain mixing and instance selection are two separate problems. |
| Approach: | They propose a framework that unifies mixing and selection without training proxy models or relying on external reference datasets. |
| Outcome: | The proposed framework achieves 2.0 data efficiency over a random baseline and further improves overall performance compared to SOTA methods in reasoning-heavy evaluations and multilingual generalization. |
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. |
Watching the Watchers: Exposing Gender Disparities in Machine Translation Quality Estimation (2025.acl-long)
Copied to clipboard
| Challenge: | Qualitative estimation (QE) metrics have been optimized to align with human quality judgments, but whether they encode social biases has been largely overlooked. |
| Approach: | They define and investigate gender bias of QE metrics and discuss its downstream implications for machine translation (MT) when a human entity’s gender in the source is undisclosed, masculine-inflected translations score higher than feminine-infflectes translations are penalized. |
| Outcome: | The proposed measures are based on gender-based quality estimation metrics across multiple domains, datasets, and languages. |
Demystifying Data Organization for Enhanced LLM Training (2026.acl-long)
Copied to clipboard
Yalun Dai, Yangyu Huang, Tongshen Yang, Yonghan Wang, Xin Zhang, Wenshan Wu, Qihao Zhao, Hao Li, Yuanyuan Gao, Kim-Hui Yap, Scarlett Li
| Challenge: | Large Language Models (LLMs) have revolutionized various fields, yet their training efficiency is heavily reliant on effective data curation. |
| Approach: | They propose to reuse pre-computed sample-level scores originally generated for data efficiency and introduce two new data ordering methods to improve LLM training. |
| Outcome: | The proposed methods improve the stability and performance of LLM training. |
Improving the Language Understanding Capabilities of Large Language Models Using Reinforcement Learning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Instruction-fine-tuned large language models (LLMs) under 14B parameters underperform on NLU tasks . we explore a framework to improve the NLU capabilities of LLMs . |
| Approach: | They propose to use Proximal Policy Optimization to improve NLU capabilities . they frame NLU as a reinforcement learning environment and optimize for reward signals . |
| Outcome: | The proposed framework outperforms supervised fine-tuning on GLUE and superGLUE tasks. |
Machine-generated text detection prevents language model collapse (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly prevalent across the web, resulting in a degenerative process whereby LLMs reinforce their own errors and reduce output diversity. |
| Approach: | They propose to use machine-generated text to reduce model collapse by up-sampling likely human content in training data. |
| Outcome: | The proposed approach prevents model collapse and improves performance compared to training on purely human data. |
-Stance: A Large-Scale Real World Dataset of Stances in Legal Argumentation (2025.acl-long)
Copied to clipboard
| Challenge: | Current tools for legal argument reasoning do not support this task. |
| Approach: | They propose to use a large-scale dataset to facilitate work on the legal argument stance classification task by evaluating whether a case summary strengthens or weakens a legal argument. |
| Outcome: | The proposed dataset is used to facilitate work on the legal argument stance classification task, which involves assessing whether a case summary strengthens or weakens a legal argument (polarity) and to what extent (intensity). |
Prior Prompt Engineering for Reinforcement Fine-Tuning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have focused on algorithms, reward shaping, and data curation, but prior prompt engineering is understudied. |
| Approach: | They investigate prior prompt engineering (pPE) in reinforcement fine-tuning . they translate five representative iPE strategies into corresponding pPE approaches . |
| Outcome: | The proposed approaches outperform iPE-prompted models on in-domain and out-of-domain benchmarks. |
Design Choices for Extending the Context Length of Visual Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Existing open-source Visual Language Models lack systematic exploration into extending their context length, and commercial models often provide limited details. |
| Approach: | They propose to extend Visual Language Models (VLMs) to 128K lengths and open-source the code, data, and models. |
| Outcome: | The proposed model is based on the Qwen-VL series model and is competitive with commercial model GPT-4V. |