Papers by Radu Florian
A Multilingual Reading Comprehension System for more than 100 Languages (2020.coling-demos)
Copied to clipboard
Anthony Ferritto, Sara Rosenthal, Mihaela Bornea, Kazi Hasan, Rishav Chakravarti, Salim Roukos, Radu Florian, Avi Sil
| Challenge: | Recent advances in open domain question answering (QA) have focused on machine reading comprehension (MRC) |
| Approach: | They propose a multilingual machine reading comprehension (MRC) demo which can answer questions in over 100 languages. |
| Outcome: | The proposed system can answer questions in over 100 languages and integrates with IBM Watson's machine translation widget to improve language accessibility. |
VAULT: VAriable Unified Long Text Representation for Machine Reading Comprehension (2021.acl-short)
Copied to clipboard
| Challenge: | Existing models on Machine Reading Comprehension (MRC) require complex model architecture for effectively modeling long texts with paragraph representation and classification, making inference computationally inefficient for production use. |
| Approach: | They propose a novel Gaussian distribution-based paragraph representation for Machine Reading Comprehension (MRC) that is light-weight and parallel-efficient. |
| Outcome: | The proposed model can achieve comparable performance on Wikipedia-based (NQ) and TechNotes (TechQA) with a state-of-the-art (SOTA) complex document modeling approach while being 16 times faster, demonstrating the efficiency of the proposed model. |
The TechQA Dataset (2020.acl-main)
Copied to clipboard
Vittorio Castelli, Rishav Chakravarti, Saswati Dana, Anthony Ferritto, Radu Florian, Martin Franz, Dinesh Garg, Dinesh Khandelwal, Scott McCarley, Michael McCawley, Mohamed Nasr, Lin Pan, Cezar Pendus, John Pitrelli, Saurabh Pujar, Salim Roukos, Andrzej Sakrajda, Avi Sil, Rosario Uceda-Sosa, Todd Ward, Rong Zhang
| Challenge: | TECHQA is a domain-adaptation question answering dataset for the technical support domain. |
| Approach: | They propose a domain-adaptation question-answering dataset for the technical support domain that contains actual questions posed by users on a technical forum . |
| Outcome: | The TECHQA dataset highlights two real-world issues from the automated customer support domain. |
Bootstrapping Multilingual AMR with Contextual Word Alignments (2021.eacl-main)
Copied to clipboard
Janaki Sheth, Young-Suk Lee, Ramón Fernandez Astudillo, Tahira Naseem, Radu Florian, Salim Roukos, Todd Ward
| Challenge: | Abstract Meaning Representation (AMR) is a sentence-level graph that is biased towards English. |
| Approach: | They propose a technique for foreign-text-to-English AMR alignment using contextual word alignment between English and foreign language tokens. |
| Outcome: | The proposed technique outperforms the best results for German, Italian, Spanish and Chinese. |
Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning (P19-1)
Copied to clipboard
| Challenge: | Abstract meaning representations (AMRs) are labeled directed acyclic graphs that represent a non intersentential abstraction of natural language with broad-coverage semantic representations. |
| Approach: | They build upon a transition-based AMR parser that uses Stack-LSTMs and augment training with policy learning. |
| Outcome: | The proposed parser performs comparable to the best published parsers. |
Towards building a Robust Industry-scale Question Answering System (2020.coling-industry)
Copied to clipboard
| Challenge: | Existing systems that use “zero-shot transfer learning” (ZSTL) are difficult to train and have observation biases. |
| Approach: | They propose a production model called GAAMA which has two characteristics . it is robust and efficient, and trains on the recently introduced Natural Questions dataset . |
| Outcome: | The proposed model performs on two benchmarks: BioASQ and CovidQA. |
CFO: A Framework for Building Production NLP Systems (D19-3)
Copied to clipboard
Rishav Chakravarti, Cezar Pendus, Andrzej Sakrajda, Anthony Ferritto, Lin Pan, Michael Glass, Vittorio Castelli, J. William Murdock, Radu Florian, Salim Roukos, Avi Sil
| Challenge: | Using a new orchestration framework, we build, test, and deploy interactive NLP and IR systems to production environments. |
| Approach: | They introduce a new orchestration framework for building, experimenting with, and deploying interactive NLP and IR systems to production environments. |
| Outcome: | The proposed framework is well suited to a variety of use cases but is not suitable for academic benchmarking or industry specific use cases. |
Pushing the Limits of AMR Parsing with Self-Learning (2020.findings-emnlp)
Copied to clipboard
Young-Suk Lee, Ramón Fernandez Astudillo, Tahira Naseem, Revanth Gangi Reddy, Radu Florian, Salim Roukos
| Challenge: | Abstract Meaning Representation (AMR) parsing has experienced a notable growth in performance in the last two years due to the impact of transfer learning and the development of novel architectures specific to AMR. |
| Approach: | They propose to use AMR annotations to generate synthetic text and refine actions oracle without additional human annotations for AMR parsing. |
| Outcome: | The proposed models improve on AMR 1.0 and 2.0 without human annotations. |
PrimeQA: The Prime Repository for State-of-the-Art Multilingual Question Answering Research and Development (2023.acl-demo)
Copied to clipboard
Avi Sil, Jaydeep Sen, Bhavani Iyer, Martin Franz, Kshitij Fadnis, Mihaela Bornea, Sara Rosenthal, Scott McCarley, Rong Zhang, Vishwajeet Kumar, Yulong Li, Md Arafat Sultan, Riyaz Bhat, Juergen Bross, Radu Florian, Salim Roukos
| Challenge: | Question Answering (QA) is a major area of research in Natural Language Processing (NLP) |
| Approach: | They propose a one-stop and open-source QA repository for question answering . it supports core QA functionalities like retrieval and reading comprehension . they say it will facilitate easy replication of state-of-the-art (SOTA) QA methods . |
| Outcome: | The proposed framework enables easy replication of state-of-the-art (SOTA) QA methods. |
Ensemble-Instruct: Instruction Tuning Data Generation with a Heterogeneous Mixture of LMs (2023.findings-emnlp)
Copied to clipboard
Young-Suk Lee, Md Sultan, Yousef El-Kurdi, Tahira Naseem, Asim Munawar, Radu Florian, Salim Roukos, Ramón Astudillo
| Challenge: | Empirical studies with different instruction-tuned LMs show that our proposed method yields higher-quality instruction tuning data than Self-Instruct. |
| Approach: | They propose to use in-context learning techniques to train strong conversational agents . they propose to categorize and simplify ICL templates to make prompt learning easier . |
| Outcome: | Empirical results show that the proposed method yields higher-quality instruction tuning data than Self-Instruct and improves performance of both vanilla and instruction-tuned LMs. |
DocAMR: Multi-Sentence AMR Representation and Evaluation (2022.naacl-main)
Copied to clipboard
Tahira Naseem, Austin Blodgett, Sadhana Kumaravel, Tim O’Gorman, Young-Suk Lee, Jeffrey Flanigan, Ramón Astudillo, Radu Florian, Salim Roukos, Nathan Schneider
| Challenge: | Abstract Meaning Representation (AMR) graphs are compared to gold graphs by the Smatch metric, but lack a well-defined representation and evaluation. |
| Approach: | They propose an algorithm for deriving a unified graph representation using a super-sentential annotation method. |
| Outcome: | The proposed algorithm avoids the pitfalls of over-merging and lacks coherence from under merging. |
Neural Cross-Lingual Relation Extraction Based on Bilingual Word Embedding Mapping (D19-1)
Copied to clipboard
| Challenge: | Relation extraction (RE) is an important information extraction task that seeks to detect and classify semantic relationships between entities. |
| Approach: | They propose a bilingual word embedding mapping approach for cross-lingual RE model transfer . they use a small bilingual dictionary with only 1K word pairs to embed word pairs . |
| Outcome: | The proposed approach achieves very good performance on target and target languages . it uses bilingual word embedding mapping to transfer a source-language model . |
ARES: A Reading Comprehension Ensembling Service (2020.emnlp-demos)
Copied to clipboard
Anthony Ferritto, Lin Pan, Rishav Chakravarti, Salim Roukos, Radu Florian, J. William Murdock, Avi Sil
| Challenge: | ARES is a machine reading comprehension (MRC) demonstration system which utilizes an ensemble of models to increase F1 by 2.3 points. |
| Approach: | They propose a machine reading comprehension (MRC) demonstration system which utilizes an ensemble of models to increase F1 by 2.3 points. |
| Outcome: | The proposed system increases F1 by 2.3 points on a short answer task using an ensemble of models. |
AMR Parsing with Action-Pointer Transformer (2021.naacl-main)
Copied to clipboard
| Challenge: | Abstract Meaning Representation parsing is a sentence-to-graph prediction task . graph nodes are semantically based on one or more sentence tokens, so implicit alignments can be derived. |
| Approach: | They propose a transition-based system that decouples hard-attention over sentences with a target-side action pointer mechanism to decouple source tokens from node representations and address alignments. |
| Outcome: | The proposed system achieves the second best Smatch score on AMR 2.0 (81.8) it decouples source tokens from node representations and addresses alignments, but lacks expressiveness. |
Scalable Cross-lingual Treebank Synthesis for Improved Production Dependency Parsers (2020.coling-industry)
Copied to clipboard
| Challenge: | scalable Universal Dependency (UD) treebank synthesis techniques are used to improve production-grade parsers. |
| Approach: | They propose a data augmentation technique that uses synthetic treebanks to improve production-grade parsers. |
| Outcome: | The proposed technique improves LAS performance on seven languages by up to two points on production models trained on original UD treebanks. |
Moving Beyond Downstream Task Accuracy for Information Retrieval Benchmarking (2023.findings-acl)
Copied to clipboard
Keshav Santhanam, Jon Saad-Falcon, Martin Franz, Omar Khattab, Avi Sil, Radu Florian, Md Arafat Sultan, Salim Roukos, Matei Zaharia, Christopher Potts
| Challenge: | Neural information retrieval (IR) systems have progressed rapidly in recent years . many IR benchmarks focus on downstream task accuracy, concealing costs incurred . |
| Approach: | They propose to include efficiency considerations on IR benchmarks to help drive progress . eral et al. propose to incorporate query latency and cost budgets into evaluation . |
| Outcome: | a new study shows that the best IR system varies according to how efficiency considerations are chosen and weighed . the proposed benchmarks would allow for more thorough exploration of possible system designs . |
Transition-based Parsing with Stack-Transformers (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing parsing systems use local or global models of the parser state to improve performance. |
| Approach: | They propose to modify the sequence-to-sequence Transformer to model global or local parser states in transition-based parsing. |
| Outcome: | The proposed model significantly improves performance on dependency and Abstract Meaning Representation (AMR) parsing tasks. |
Neural Cross-Lingual Coreference Resolution And Its Application To Entity Linking (P18-2)
Copied to clipboard
| Challenge: | a cross-lingual coreference model is based on multi-lingual embeddings and language independent features. |
| Approach: | They propose a crosslingual coreference model that builds on multi-lingual embeddings and language independent features. |
| Outcome: | The proposed model outperforms the existing models on Chinese and Spanish test sets. |
Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to limit overfitting of training domains are rooted in this problem . domain generalization (DG) seeks to train models on a small number of source domains . |
| Approach: | They propose to use knowledge distillation to train models on a small number of source domains to maximize their zero-shot out-of-domain utility. |
| Outcome: | The proposed model learns its source domains better and has better out-of-domain generalization . the proposed model outperforms existing approaches that aim to limit overfitting . |
Structure-aware Fine-tuning of Sequence-to-sequence Transformers for Transition-based AMR Parsing (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent work shows that pre-trained sequence-to-sequence Transformer models are effective in predicting linearized Abstract Meaning Representation graphs. |
| Approach: | They propose a structure-aware transition-based approach to AMR parsing that integrates general pre-trained sequence-to-sequence language models with a structured transition set. |
| Outcome: | The proposed approach retains the desirable properties of previous approaches while reaching the new parsing state of the art for AMR 2.0. |
From Multiple-Choice to Extractive QA: A Case Study for English and Arabic (2025.coling-main)
Copied to clipboard
Teresa Lynn, Malik H. Altakrori, Samar M. Magdy, Rocktim Jyoti Das, Chenyang Lyu, Mohamed Nasr, Younes Samih, Kirill Chirkunov, Alham Fikri Aji, Preslav Nakov, Shantanu Godbole, Salim Roukos, Radu Florian, Nizar Habash
| Challenge: | Recent years have brought about very fast developments in Natural Language Processing (NLP), but many other languages are overlooked due to limited resources. |
| Approach: | They propose to repurpose a multilingual BELEBELE dataset for a task of extractive QA in the style of machine reading comprehension. |
| Outcome: | The proposed approach could be used to extract QA in the style of machine reading comprehension. |
UDAPDR: Unsupervised Domain Adaptation via LLM Prompting and Distillation of Rerankers (2023.emnlp-main)
Copied to clipboard
Jon Saad-Falcon, Omar Khattab, Keshav Santhanam, Radu Florian, Martin Franz, Salim Roukos, Avirup Sil, Md Sultan, Christopher Potts
| Challenge: | Existing methods for information retrieval tasks require large labeled datasets for fine-tuning, but they can experience significant drops in accuracy due to distribution shifts from the training to the target domain. |
| Approach: | They propose a method for using large language models to generate large numbers of synthetic queries cheaply using an expensive LLM. |
| Outcome: | The proposed method boosts zero-shot accuracy in long-tail domains and achieves substantially lower latency than standard reranking methods. |
GPT-too: A Language-Model-First Approach for AMR-to-Text Generation (2020.acl-main)
Copied to clipboard
Manuel Mager, Ramón Fernandez Astudillo, Tahira Naseem, Md Arafat Sultan, Young-Suk Lee, Radu Florian, Salim Roukos
| Challenge: | Existing approaches to generating text from AMRs focus on training sequence-to-sequence or graph-tosequent models on annotated data. |
| Approach: | They propose a strong pre-trained language model with cycle consistency-based re-scoring to generate AMR text. |
| Outcome: | The proposed model outperforms existing methods on the English LDC2017T10 dataset. |
Maximum Bayes Smatch Ensemble Distillation for AMR Parsing (2022.naacl-main)
Copied to clipboard
| Challenge: | AMR parsing has experienced an unprecendented increase in performance in the last three years due to a mixture of effects including architecture improvements and transfer learning. |
| Approach: | They propose to combine Smatch-based ensembling techniques with ensemble distillation to overcome this diminishing returns of silver data. |
| Outcome: | The proposed technique can produce gains rivaling those of human annotated data for QALD-9 and achieve a new state-of-the-art for BioAMR. |
A Grounded Preference Model for LLM Alignment (2024.findings-acl)
Copied to clipboard
Tahira Naseem, Guangxuan Xu, Sarathkrishna Swaminathan, Asaf Yehudai, Subhajit Chaudhury, Radu Florian, Ramón Astudillo, Asim Munawar
| Challenge: | Large Language Models (LLMs) suffer from factual inconsistency and hallucination despite recent advances . training a preference model requires substantial human annotation, which is expensive and labor-intensive. |
| Approach: | They propose to generate synthetic grounded preference data and train a Grounded Preference Model to assess the overall quality of grounded responses. |
| Outcome: | The proposed model can generate much better grounded responses as judged by GPT4 and achieves the TRUE faithfulness Benchmark. |
CLAPnq: Cohesive Long-form Answers from Passages in Natural Questions for RAG systems (2025.tacl-1)
Copied to clipboard
| Challenge: | Retrieval Augmented Generation (RAG) is a popular application for large language models. |
| Approach: | They present a benchmark Long-form Question Answering dataset for RAG . they use a passage grounded in a gold passage to provide accurate answers . |
| Outcome: | The proposed dataset provides long answers with grounded gold passages from Natural Questions (NQ) the answers are concise, 3x smaller than the full passage, and cohesive, meaning the answer is composed fluently. |
Centurio: On Drivers of Multilingual Ability of Large Vision-Language Model (2025.acl-long)
Copied to clipboard
Gregor Geigle, Florian Schneider, Carolin Holtermann, Chris Biemann, Radu Timofte, Anne Lauscher, Goran Glavaš
| Challenge: | Existing models for large vision-language tasks are trained on English data, which makes them struggle to understand non-English input and fail to generate output in the desired target language. |
| Approach: | They conduct multi-stage experiments on 13 vision-language tasks and 43 languages . they find that one can include as many as 100 training languages simultaneously with as little as 25-50% of non-English data . |
| Outcome: | The proposed model outperforms existing models in 14 tasks and 56 languages. |
Inducing and Using Alignments for Transition-based AMR Parsing (2022.naacl-main)
Copied to clipboard
Andrew Drozdov, Jiawei Zhou, Radu Florian, Andrew McCallum, Tahira Naseem, Yoon Kim, Ramón Astudillo
| Challenge: | Abstract Meaning Representation parsers rely on node-to-word alignments, but lack the complexity of the pipeline. |
| Approach: | They propose a neural aligner for abstract meaning representation that learns node-to-word alignments without relying on pipelines. |
| Outcome: | The proposed approach improves accuracy and generalization from AMR2.0 to AMR3.0 corpora. |
Multi-Stage Pre-training for Low-Resource Domain Adaptation (2020.emnlp-main)
Copied to clipboard
Rong Zhang, Revanth Gangi Reddy, Md Arafat Sultan, Vittorio Castelli, Anthony Ferritto, Radu Florian, Efsun Sarioglu Kayi, Salim Roukos, Avi Sil, Todd Ward
| Challenge: | Existing approaches to transfer learning target data to in-domain text . prior work has adapted pre-trained LMs to specific domains . |
| Approach: | They extend the vocabulary of a pretrained language model with domain-specific terms to create synthetic tasks that help it transfer to downstream tasks. |
| Outcome: | The proposed approaches show significant performance gains on extractive reading comprehension, document ranking and duplicate question detection tasks. |