Papers by Gholamreza Haffari

Uncertainty-Aware Balancing for Multilingual and Multi-Domain Neural Machine Translation Training (2021.emnlp-main)

Challenge:	Existing distributional alignment models are unstable and degrade under cultural and domain shifts.
Approach:	They propose a distributional alignment technique that improves distribution prediction under cultural and domain shift.
Outcome:	The proposed method improves fidelity and robustness of LLM distribution estimation under domain and cultural shift.

Copied to clipboard

Minghao Wu, Yitong Li, Meng Zhang, Liangyou Li, Gholamreza Haffari, Qun Liu

Presentation Matters: How to Communicate Science in the NLP Venues and in the Wild? (2024.acl-tutorials)

Challenge:	MultiUAT dynamically adjusts training data usage based on model’s uncertainty on a small set of trusted clean data for multi-corpus machine translation.
Approach:	They propose an approach that dynamically adjusts the training data usage based on the model’s uncertainty on a small set of trusted clean data for multi-corpus machine translation.
Outcome:	The proposed approach outperforms baselines on 16 languages and 2 domains on English-German translation.

Copied to clipboard

Sarvnaz Karimi, Cecile Paris, Gholamreza Haffari

Challenge:	a tutorial on communication skills is being proposed to help early career researchers . the tutorial would cover writing, oral presentation and social media presence .
Approach:	a tutorial on communication skills is proposed to help early career researchers . the tutorial would cover writing, oral presentation and social media presence .
Outcome:	a new tutorial will cover communication skills, including writing, oral presentation and social media presence . the tutorial will allow attendees to ask questions and clarify their research .

Towards relation extraction from speech (2022.emnlp-main)

Copied to clipboard

Tongtong Wu, Guitao Wang, Jinming Zhao, Zhaoran Liu, Guilin Qi, Yuan-Fang Li, Gholamreza Haffari

Multimodality

Challenge:	Existing methods for extracting relations from speech have been neglected due to the nature of speech.
Approach:	They propose a listening information extraction task that uses speech to extract relation extraction from speech . they use a text-to-speech system and crowd-sourced native English speakers to test the task .
Outcome:	The proposed task extracts semantic relationships from speech data using a new model . the proposed task is more challenging than the existing method due to the characteristics of speech .

NAP2: A Benchmark for Naturalness and Privacy-Preserving Text Rewriting by Learning from Human (2025.findings-emnlp)

Copied to clipboard

Shuo Huang, William Maclean, Xiaoxi Kang, Qiongkai Xu, Zhuang Li, Xingliang Yuan, Gholamreza Haffari, Lizhen Qu

SG-FSM: A Self-Guiding Zero-Shot Prompting Paradigm for Multi-Hop Question Answering Based on Finite State Machine (2025.findings-naacl)

Challenge:	a large number of large language models are being used to protect user privacy . sanitizing sensitive text using two common strategies is the answer .
Approach:	They propose sanitizing sensitive text using deleting expressions and abstracting them . they propose a tool for text rewriting that uses crowdsourcing and large language models .
Outcome:	The proposed approach protects privacy before sending sensitive data to large language models . it combines crowdsourcing and large language modeling to create a text rewrite tool .

Copied to clipboard

Xiaochen Wang, Junqing He, Liang Chen, Gholamreza Haffari, Yiru Wang, Zhe Yang, Xiangdi Meng, Kunhao Pan, Zhifang Sui

Document Flattening: Beyond Concatenating Context for Document-Level Neural Machine Translation (2023.eacl-main)

Challenge:	Multi-hop Question Answering (MHQA) is a challenging task that requires models to answer multiple questions with multiple passages.
Approach:	They propose a self-guided prompting finite state machine to improve multi-hop reasoning abilities by iterating over multiple questions and correcting itself to improve accuracy.
Outcome:	The proposed approach outperforms baselines on Musique and other datasets.

Copied to clipboard

Minghao Wu, George Foster, Lizhen Qu, Gholamreza Haffari

Improving Numerical Reasoning Skills in the Modular Approach for Complex Question Answering on Text (2021.findings-emnlp)

Challenge:	Existing document-level neural machine translation systems concatenate several consecutive sentences to form a pseudo-document, and then learn inter-sentential dependencies.
Approach:	They propose a document flattening technique that integrates Flat-Batch Attention (FBA) and Neural Context Gate (NCG) into Transformer model to utilize information beyond the pseudo-document boundaries.
Outcome:	The proposed method outperforms baselines on BLEU, COMET and accuracy on the contrastive test set.

Copied to clipboard

Xiao-Yu Guo, Yuan-Fang Li, Gholamreza Haffari

The Context-Dependent Additive Recurrent Neural Net (N18-1)

Challenge:	Neural Module Networks (NMNs) is an end-to-end differentiable model in the programmer-interpreter paradigm.
Approach:	They propose to make the interpreter question-aware and capture the relationship between entities and numbers in both questions and paragraphs.
Outcome:	The proposed models outperform the original models on the DROP dataset and are interpertable by nature.

Copied to clipboard

Quan Hung Tran, Tuan Lai, Gholamreza Haffari, Ingrid Zukerman, Trung Bui, Hung Bui

Natural Language Interfaces

Challenge:	Contextual sequence mapping is one of the fundamental problems in Natural Language Processing (NLP).
Approach:	They propose a new family of Recurrent Neural Networks that address contextual sequence mapping . they propose to use contextual signals to control the flow of information .
Outcome:	The proposed architecture outperforms existing methods on dialog problem and language model . the proposed architectures are based on a novel family of recurrent neural networks .

It Is Not As Good As You Think! Evaluating Simultaneous Machine Translation on Interpretation Data (2021.emnlp-main)

Copied to clipboard

Jinming Zhao, Philip Arthur, Gholamreza Haffari, Trevor Cohn, Ehsan Shareghi

MADial-Bench: Towards Real-world Evaluation of Memory-Augmented Dialogue Generation (2025.naacl-long)

Challenge:	Existing siMT systems are trained and evaluated on offline translations . however, evaluation gap remains notable, calling for constructing large-scale interpretation corpora .
Approach:	They propose a translation-to-interpretation transfer method which converts offline translations into interpretation-style data.
Outcome:	The proposed interpretation test set shows that SiMT models improve on translation vs interpretation data.

Copied to clipboard

Junqing He, Liang Zhu, Rui Wang, Xi Wang, Gholamreza Haffari, Jiaxing Zhang

Neural Speech Translation using Lattice Transformations and Graph Networks (D19-53)

Challenge:	Existing evaluation metrics for memory-augmented dialogue systems lack practical value . current evaluation methods only consider passive memory retrieval while ignoring diverse memory recall with rich triggering factors.
Approach:	They propose to use long-term memory to create human-like dialogues using chatbots.
Outcome:	The proposed benchmark covers memory retrieval and memory recognition tasks with both passive and proactive memory recall data.

Copied to clipboard

Daniel Beck, Trevor Cohn, Gholamreza Haffari

Multimodality

Challenge:	Existing work on end-to-end systems bypass the need for intermediate representations, but this approach is limited in practical applications.
Approach:	They propose a lattice-tosequence model which uses lattics as encoders and graph networks to address two problems by applying latticae transformations and a neural model.
Outcome:	The proposed model beats pipeline approaches while being orders of magnitude faster than previous work.

Generating Synthetic Speech from SpokenVocab for Speech Translation (2023.findings-eacl)

Copied to clipboard

Jinming Zhao, Gholamreza Haffari, Ehsan Shareghi

T3L: Translate-and-Test Transfer Learning for Cross-Lingual Text Classification (2023.tacl-1)

Challenge:	End-to-end speech-totext translation (ST) models require large amounts of data to train, but their size is considerably smaller than text-based MT data.
Approach:	They propose a method to convert MT data to ST data via text-to-speech systems.
Outcome:	The proposed method improves translation quality by an average of 1.83 BLEU score while performing equally well as TTS-generated speech in improving translation quality.

Copied to clipboard

Inigo Jauregi Unanue, Gholamreza Haffari, Massimo Piccardi

Extending LLMs to New Languages: A Case Study of Llama and Persian Adaptation (2025.coling-main)

Challenge:	Existing approaches to cross-lingual text classification leverage text classifiers trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning.
Approach:	They propose to combine a neural machine translator and a text classifier trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning.
Outcome:	The proposed approach significantly improves over a baseline approach.

Copied to clipboard

Samin Mahdizadeh Sani, Pouya Sadeghi, Thuy-Trang Vu, Yadollah Yaghoobzadeh, Gholamreza Haffari

Information Retrieval

Challenge:	Large language models (LLMs) are mainly trained on English data and struggle with low-resource languages.
Approach:	They propose to add a new language to Llama to improve classification accuracy for Persian tasks by aligning representations through bilingual pretraining and instruction datasets.
Outcome:	The proposed model performs on generation and classification tasks with no adverse impact and sometimes even improvements on English tasks.

Document Context Neural Machine Translation with Memory Networks (P18-1)

Copied to clipboard

Sameen Maruf, Gholamreza Haffari

RedApt: An Adaptor for wav2vec 2 EncodingFaster and Smaller Speech Translation without Quality Compromise (2022.findings-emnlp)

Challenge:	Experimental results show that our model exploits both source and target document context.
Approach:	They propose a document-level neural machine translation model which takes both source and target document context into account using memory networks.
Outcome:	The proposed model outperforms previous work in terms of BLEU and METEOR in English translations.

Copied to clipboard

Jinming Zhao, Hao Yang, Gholamreza Haffari, Ehsan Shareghi

Speech & Audio in NLP

Challenge:	Pre-trained speech Transformers in speech translation systems have facilitated state-of-the-art (SotA) results, but their computational cost is high.
Approach:	They propose a Reducer Adaptor block that could be seamlessly integrated within any Transformer-based speech encoding architecture.
Outcome:	The proposed Reducer Adaptor block outperforms the existing SotA architecture by an average of 0.68 BLEU score on 8 language pairs from Must-C.

Automatic Post-Editing of Machine Translation: A Neural Programmer-Interpreter Approach (D18-1)

Copied to clipboard

Thuy-Trang Vu, Gholamreza Haffari

Multilingual Neural Machine Translation: Can Linguistic Hierarchies Help? (2021.findings-emnlp)

Challenge:	Existing approaches to inducing APE have suffered from over-correction, where the APE system tends to keep the machine translated text without any modification.
Approach:	They propose a neural programmer-interpreter approach to automated post-editing (APE) that mimics human perform post- editing using discrete edit operations . their model outperforms previous neural models for inducing PE programs on the WMT17 APE task for German-English up to +1 BLEU score and -0.7 TER scores.
Outcome:	The proposed model outperforms previous neural models for inducing PE programs on the WMT17 APE task for German-English up to +1 BLEU score and -0.7 TER scores.

Copied to clipboard

Fahimeh Saleh, Wray Buntine, Gholamreza Haffari, Lan Du

Student Surpasses Teacher: Imitation Attack for Black-Box NLP APIs (2022.coling-1)

Challenge:	Multilingual Neural Machine Translation (MNMT) trains a single model that supports translation between multiple languages . transferring knowledge from a diverse set of languages degrades the translation performance due to negative transfer.
Approach:	They propose a hierarchical knowledge distillation approach to train multilingual models . they use typological features and phylogeny to overcome negative transfer issue .
Outcome:	The proposed approach avoids negative transfer effect by capitalising on language groups generated according to typological features and phylogeny of languages.

Copied to clipboard

Qiongkai Xu, Xuanli He, Lingjuan Lyu, Lizhen Qu, Gholamreza Haffari

Dynamic Programming Encoding for Subword Segmentation in Neural Machine Translation (2020.acl-main)

Challenge:	Existing MLaaS models are vulnerable to imitation attacks, but none of the stolen models can outperform the original black-box APIs.
Approach:	They conduct unsupervised domain adaptation and multi-victim ensemble to show attackers could surpass victims.
Outcome:	The proposed model outperforms the original black-box models on transferred domains.

Copied to clipboard

Xuanli He, Gholamreza Haffari, Mohammad Norouzi

Selective Attention for Context-aware Neural Machine Translation (N19-1)

Challenge:	Empirical results on machine translation suggest that DPE is effective for segmenting output sentences.
Approach:	They propose a new algorithm for tokenizing sentences into subword units . they propose enabling exact log marginal likelihood estimation and exact MAP inference .
Outcome:	The proposed algorithm improves on machine translation datasets and on a large dataset.

Copied to clipboard

Sameen Maruf, André F. T. Martins, Gholamreza Haffari

On Robustness of Prompt-based Semantic Parsing with Large Pre-trained Language Model: An Empirical Study on Codex (2023.eacl-main)

Challenge:	Recent work in context-aware NMT considers only a few previous sentences as context . current systems fail to achieve fluent, good quality translation for a full document .
Approach:	They propose a top-down approach to hierarchical attention for context-aware NMT which uses sparse attention to selectively focus on relevant sentences in the document context.
Outcome:	The proposed approach outperforms context-agnostic baselines and context-based baselines on English-German datasets.

Copied to clipboard

Terry Yue Zhuo, Zhuang Li, Yujin Huang, Fatemeh Shiri, Weiqing Wang, Gholamreza Haffari, Yuan-Fang Li

NormMark: A Weakly Supervised Markov Model for Socio-cultural Norm Discovery (2023.findings-acl)

Challenge:	Existing techniques for parsing natural-language utterances are vulnerable to adversarial attacks, requiring large amounts of labelled data and expensive human annotation.
Approach:	They propose to enhance the adversarial robustness of a prompt-based semantic parser based on a language model trained on code by constructing a set of demonstration examples.
Outcome:	The proposed method can be enhanced without significant amounts of labelled data or expensive human annotations on in-domain semantic parsing data.

Copied to clipboard

Farhad Moghimifar, Shilin Qu, Tongtong Wu, Yuan-Fang Li, Gholamreza Haffari

Adaptive Knowledge-Enhanced Bayesian Meta-Learning for Few-shot Event Detection (2021.findings-acl)

Challenge:	Existing methods for norm recognition focus only on surface-level features of dialogues and do not take into account the interactions within a conversation.
Approach:	They propose a probabilistic generative Markov model to carry latent features throughout a dialogue and trainable on weakly annotated data using the variational technique.
Outcome:	The proposed model outperforms current state-of-the-art methods on a weakly annotated dataset, outperforming existing methods, including GPT3.

Copied to clipboard

Shirong Shen, Tongtong Wu, Guilin Qi, Yuan-Fang Li, Gholamreza Haffari, Sheng Bi

Challenge Dataset of Cognates and False Friend Pairs from Indian Languages (2020.lrec-1)

Challenge:	Event detection typically does not have sufficient labelled data, thus can be formulated as a few-shot learning problem.
Approach:	They propose a knowledge-based fewshot event detection method which introduces external event knowledge as the knowledge prior of event types.
Outcome:	Experiments show that the proposed method outperforms baselines by 15 F 1 points . event detection is an important task in information extraction .

Copied to clipboard

Diptesh Kanojia, Malhar Kulkarni, Pushpak Bhattacharyya, Gholamreza Haffari

On the Reliability of Large Language Models for Causal Discovery (2025.acl-long)

Challenge:	Cognates are words that have a common etymological origin and can facilitate the Second Language Acquisition (SLA) however, they also pose a challenge to various NLP applications such as Machine Translation and Cross-lingual Sense Disambiguation.
Approach:	They create two cognate datasets for twelve Indian languages and use them to generate cognate sets.
Outcome:	The proposed datasets are curated using previously available baseline cognate detection approaches and evaluated with the help of lexicographers.

Copied to clipboard

Tao Feng, Lizhen Qu, Niket Tandon, Zhuang Li, Xiaoxi Kang, Gholamreza Haffari

Variational Autoencoder with Disentanglement Priors for Low-Resource Task-Specific Natural Language Generation (2022.emnlp-main)

Challenge:	Existing statistical methods to identify causal relationships from observational data remain elusive.
Approach:	They examine the impact of memorization for accurate causal relation prediction, the influence of incorrect causal relations in pre-training data and the contextual nuances that influence LLMs’ understanding of causal relations.
Outcome:	The proposed models are effective in recognizing causal relations that occur frequently in pre-training data, but their ability to generalize to new or rare causal relations is limited.

Copied to clipboard

Zhuang Li, Lizhen Qu, Qiongkai Xu, Tongtong Wu, Tianyang Zhan, Gholamreza Haffari

Low-Resource NLP

Challenge:	Existing models for task-specific natural language generation do not contain any labeled examples.
Approach:	They propose a variational autoencoder with disentanglement priors for task-specific natural language generation with none or a handful of task-related labeled examples.
Outcome:	The proposed model outperforms baseline models in terms of data augmentation and text style transfer in the few-shot setting.

Active Learning for Multilingual Semantic Parser (2023.findings-eacl)

Copied to clipboard

Zhuang Li, Gholamreza Haffari

Discrete Minds in a Continuous World: Do Language Models Know Time Passes? (2025.findings-emnlp)

Challenge:	Existing multilingual semantic parsing datasets are limited in translation effort due to data imbalance.
Approach:	They propose a first active learning procedure for multilingual semantic parsing (AL-MSP) it selects only a subset from existing datasets to be translated, they propose .
Outcome:	The proposed method significantly reduces translation costs with ideal selection methods.

Copied to clipboard

Minghan Wang, Ye Bai, Thuy-Trang Vu, Ehsan Shareghi, Gholamreza Haffari

CausalScore: An Automatic Reference-Free Metric for Assessing Response Relevance in Open-Domain Dialogue Systems (2025.coling-main)

Challenge:	Large Language Models excel at temporal reasoning tasks, but their ability to perceive the passage of time remains unexplored.
Approach:	They propose a Token-Time Hypothesis to test whether LLMs perceive the passage of time . they also propose an interactive navigation challenge to examine how LLM responds to time pressure .
Outcome:	The proposed model can map discrete token counts to wall-clock time and validate this through a dialogue duration judgment task.

Copied to clipboard

Tao Feng, Lizhen Qu, Xiaoxi Kang, Gholamreza Haffari

Natural Language Interfaces

Challenge:	Existing metrics for dialogue quality evaluation show low correlation with human judgements . current metrics do not accurately evaluate dialogue responses based on dialogue history .
Approach:	They propose a new metric measuring causal strength between dialogue histories and responses . they collect a dialogue dataset with human-annotated causal relations and pairwise human judgements .
Outcome:	The proposed metric outperforms existing state-of-the-art metrics in human judgements . it is based on a dialogue dataset with human-annotated causal relations and human judgement sets .

Adaptively Scheduled Multitask Learning: The Case of Low-Resource Neural Machine Translation (D19-56)

Copied to clipboard

Poorya Zaremoodi, Gholamreza Haffari

SCAR: Data Selection via Style Consistency-Aware Response Ranking for Efficient Instruction-Tuning of Large Language Models (2025.acl-long)

Challenge:	Neural Machine Translation suffers from the lack of bilingual data in low-resource scenarios.
Approach:	They propose to inject inductive biases into Neural Machine Translation (NMT) using auxiliary syntactic and semantic tasks.
Outcome:	The proposed approach improves translation quality by reweighing training data of main and auxiliary tasks based on their contributions to generalisability of main task.

Copied to clipboard

Zhuang Li, Yuncheng Hua, Thuy-Trang Vu, Haolan Zhan, Lizhen Qu, Gholamreza Haffari

Total Recall: a Customized Continual Learning Method for Neural Semantic Parsers (2021.emnlp-main)

Challenge:	Recent studies show that manually ensuring a consistent response style and maintaining high data quality can significantly improve the performance of fine-tuned Large Language Models (LLMs).
Approach:	They introduce a style-aware response ranking system that prioritizes instruction-response pairs based on their stylistic consistency.
Outcome:	The proposed model matches or surpasses models trained on the entire dataset in coding and open-ended question-answering benchmarks.

Copied to clipboard

Zhuang Li, Lizhen Qu, Gholamreza Haffari

Multilingual Simultaneous Neural Machine Translation (2021.findings-acl)

Challenge:	Existing methods for continual learning for semantic parsing fail to account for special properties of structured outputs . retraining from scratch is not feasible due to the fast growing number of tasks .
Approach:	They propose a continual learning method that uses sequential learning to learn tasks without accessing full training data from previous tasks.
Outcome:	The proposed method achieves a 3-6 times speedup compared to re-training from scratch.

Copied to clipboard

Philip Arthur, Dongwon Ryu, Gholamreza Haffari

Leveraging Discourse Rewards for Document-Level Neural Machine Translation (2020.coling-main)

Challenge:	Simultaneous machine translation (SIMT) involves translating source utterances to the target language in real-time before the speaker utterrance completes.
Approach:	They propose a multilingual approach to simultaneous machine translation where a single model simultaneously translates between multiple languages.
Outcome:	The proposed multilingual approach improves on two Germanic and three Romance languages and is on-par or better than the universal model trained for all languages.

Copied to clipboard

Inigo Jauregi Unanue, Nazanin Esmaili, Gholamreza Haffari, Massimo Piccardi

Neural Machine Translation for Bilingually Scarce Scenarios: a Deep Multi-Task Learning Approach (N18-1)

Challenge:	Document-level machine translation models are often not trained to explicitly ensure discourse quality.
Approach:	They propose a method that explicitly optimizes lexical cohesion and coherence metrics by using a reinforcement learning objective.
Outcome:	The proposed approach improves document translations over four different languages and three translation domains while maintaining faithfulness to the reference translation.

Copied to clipboard

Poorya Zaremoodi, Gholamreza Haffari

The Best of Both Worlds: Combining Human and Machine Translations for Multilingual Semantic Parsing with Active Learning (2023.acl-long)

Challenge:	Neural machine translation requires large amount of parallel training text to learn a reasonable quality translation model.
Approach:	They propose a multi-task learning approach that leverages monolingual linguistic resources in the source side of a machine translation task.
Outcome:	The proposed approach is effective on three translation tasks: English-to-French, English- to-Farsi, and English-à-Vietnamese.

Copied to clipboard

Zhuang Li, Lizhen Qu, Philip Cohen, Raj Tumuluri, Gholamreza Haffari

SurveyPilot: an Agentic Framework for Automated Human Opinion Collection from Social Media (2025.acl-long)

Challenge:	Prior studies have focused on translating utterances from high-resource languages to low-resourced languages.
Approach:	They propose an active learning approach that exploits the strengths of both human and machine translations by iteratively adding small batches of human translations into the machine-translated training set.
Outcome:	The proposed approach reduces errors and bias in the translated data, resulting in higher parser accuracies than the current model trained on machine translations.

Copied to clipboard

Viet Thanh Pham, Lizhen Qu, Zhuang Li, Suraj Sharma, Gholamreza Haffari

Sentiment Analysis

Challenge:	Existing methods for opinion survey research exhibit severe biases and lack traceability.
Approach:	They propose a finite-state orchestrated agentic framework that automates the collection and analysis of human opinions from social media platforms.
Outcome:	The proposed framework achieves close alignment with authentic survey results across multiple domains, with average relative improvements of 68,98% and 51,37% when compared to opinion synthesis and agent-based approaches.

Cognition-aware Cognate Detection (2021.eacl-main)

Copied to clipboard

Diptesh Kanojia, Prashant Sharma, Sayali Ghodekar, Pushpak Bhattacharyya, Gholamreza Haffari, Malhar Kulkarni

IRIS: An Iterative and Integrated Framework for Verifiable Causal Discovery in the Absence of Tabular Data (2025.acl-long)

Challenge:	Existing approaches to cognate detection use orthographic, phonetic and semantic similarity based features sets.
Approach:	They propose a method for enriching feature sets with cognitive features extracted from gaze behaviour data from human readers’ gaze behaviour.
Outcome:	The proposed method improves cognate detection performance by 10% and 12% over existing methods.

Copied to clipboard

Tao Feng, Lizhen Qu, Niket Tandon, Gholamreza Haffari

Structured Data in NLP

Challenge:	Existing statistical methods for causal discovery are expensive, require high-quality structured tabular data, and are often not available for a wide range of NLP applications.
Approach:	They propose a framework that combines statistical and large language model methods to discover causal relations from a set of initial variables.
Outcome:	The proposed method combines statistical and LLM-based methods to discover known and novel causal relations.

Collective Wisdom: Improving Low-resource Neural Machine Translation using Adaptive Knowledge Distillation (2020.coling-main)

Copied to clipboard

Fahimeh Saleh, Wray Buntine, Gholamreza Haffari

Adaptive Knowledge Sharing in Multi-Task Learning: Improving Low-Resource Neural Machine Translation (P18-2)

Challenge:	Existing approaches to train high-quality NMT models in bilingually low-resource scenarios are limited by the scarcity of parallel sentence-pairs.
Approach:	They propose to distill the knowledge of teacher models to a single student model by using knowledge distillation.
Outcome:	The proposed approach achieves up to +0.9 BLEU score improvements compared to strong baselines.

Copied to clipboard

Poorya Zaremoodi, Wray Buntine, Gholamreza Haffari

Low-Resource NLP

Challenge:	Neural Machine Translation (NMT) requires large amounts of bilingual data to learn a translation model with reasonable quality.
Approach:	They propose to extend recurrent units with multiple "blocks" along with a trainable "routing network" this allows for adaptive collaboration by dynamic sharing of blocks conditioned on the task at hand, input, and model state.
Outcome:	Empirical evaluations of two low-resource translation tasks show +1 BLEU score improvements compared to strong baselines.

Learning Coupled Policies for Simultaneous Machine Translation using Imitation Learning (2021.eacl-main)

Copied to clipboard

Philip Arthur, Trevor Cohn, Gholamreza Haffari

Graph-to-Sequence Learning using Gated Graph Neural Networks (P18-1)

Challenge:	Existing approaches to learn simultaneous translation model with coupled programmer-interpreter policies are suboptimal as they fix the agent's policy to focus learning the NMT model or learn adaptive agent policies while the NRT model is fixed.
Approach:	They propose an algorithmic oracle to produce oracular READ/WRITE actions for training bilingual sentence-pairs using the notion of word alignments.
Outcome:	The proposed method outperforms baselines in terms of translation quality quality while keeping the delay low.

Copied to clipboard

Daniel Beck, Gholamreza Haffari, Trevor Cohn

Structured Data in NLP

Challenge:	Existing approaches to graph-to-sequence learning ignore the full graph structure, discarding key information.
Approach:	They propose a graph-to-sequence learning model that encodes the full graph structure and an input transformation that allows nodes and edges to have their own hidden representations.
Outcome:	The proposed model outperforms baselines in generation from AMR graphs and syntax-based neural machine translation while retaining the full graph structure.

CONGRAD: Conflicting Gradient Filtering for Multilingual Preference Alignment (2026.eacl-long)

Copied to clipboard

Jiangnan Li, Thuy-Trang Vu, Christian Herold, Amirhossein Tebbifakhr, Shahram Khadivi, Gholamreza Haffari

Teaching Neural Module Networks to Do Arithmetic (2022.coling-1)

Challenge:	Naive joint training of large language models can suffer from negative interference.
Approach:	They propose a filtering method that aggregates cross-lingually beneficial gradients and filters for those with high cross-linguistic affinity.
Outcome:	The proposed method outperforms baselines in both seen and unseen languages with minimal alignment tax.

Copied to clipboard

Jiayi Chen, Xiao-Yu Guo, Yuan-Fang Li, Gholamreza Haffari

CoV: Chain-of-View Prompting for Spatial Reasoning (2026.findings-acl)

Challenge:	Neural Module Networks (NMNs) have limited reasoning abilities and lack numerical reasoning capability.
Approach:	They propose to integrate the original question in the interpreter and introduce addition and subtraction modules that perform numerical reasoning over numbers.
Outcome:	The proposed methods outperform previous state-of-the-art models on a subset of DROP and achieve competitive reasoning performance.

Copied to clipboard

Haoyu Zhao, Akide Liu, Zeyu Zhang, Weijie Wang, Feng Chen, Ruihan Zhu, Gholamreza Haffari, Bohan Zhuang

Low-Resource NLP

Challenge:	Embodied question answering requires collecting context that is distributed across multiple viewpoints . most recent vision–language models (VLMs) are constrained to a fixed and finite set of input views .
Approach:	They propose a training-free, test-time reasoning framework that transforms a VLM into an active viewpoint reasoner through a coarse-to-fine exploration process.
Outcome:	The proposed framework improves LLM-Match performance by 11.98% on four mainstream VLMs.

LiveCultureBench: a Multi-Agent, Multi-Cultural Benchmark for Large Language Models in Dynamic Social Simulations (2026.acl-long)

Copied to clipboard

Viet Thanh Pham, Lizhen Qu, Thuy-Trang Vu, Gholamreza Haffari, Dinh Phung

Incorporating Syntactic Uncertainty in Neural Machine Translation with a Forest-to-Sequence Model (C18-1)

Challenge:	Large language models (LLMs) are increasingly deployed as autonomous agents . evaluations focus primarily on task success rather than cultural appropriateness or reliability.
Approach:	They propose a multi-cultural, dynamic benchmark that embeds large language models as agents in a simulated town and evaluates them on task completion and adherence to socio-cultural norms.
Outcome:	The proposed model evaluates LLMs on task completion and adherence to socio-cultural norms across models and cultural profiles.

Copied to clipboard

Poorya Zaremoodi, Gholamreza Haffari

Learning How to Actively Learn: A Deep Imitation Learning Approach (P18-1)

Challenge:	Incorporating syntactic information in machine translation can lead to better reorderings, especially useful when the language pairs are syntaktically highly divergent.
Approach:	They propose a forest-to-sequence NMT model which uses exponentially many parse trees of the source sentence to compensate for parser errors.
Outcome:	The proposed model outperforms the sequence-to-sequence and tree-to tree-based models on English, Chinese and Farsi translation tasks.

Copied to clipboard

Ming Liu, Wray Buntine, Gholamreza Haffari

Information Extraction & Text Mining

Challenge:	Experimental results show that heuristic-based active learning methods are limited when the data distribution of the underlying learning problems vary.
Approach:	They propose a method that learns an AL "policy" using "imitation learning" they use an efficient "algorithmic expert" which provides the policy learner with good actions in the encountered AL situations.
Outcome:	The proposed method is more effective than previous methods on two tasks . labeled data is rare while unlabelled data is abundant .

Generalised Unsupervised Domain Adaptation of Neural Machine Translation with Cross-Lingual Data Selection (2021.emnlp-main)

Copied to clipboard

Thuy-Trang Vu, Xuanli He, Dinh Phung, Gholamreza Haffari

Zero-Shot Privacy-Aware Text Rewriting via Iterative Tree Search (2025.findings-emnlp)

Challenge:	Existing work on unsupervised domain adaptation of neural machine translation assumes access to monolingual text in either the source or target language in the new domain.
Approach:	They propose a method to extract in-domain sentences from a large generic monolingual corpus from 'missing' text.
Outcome:	The proposed method outperforms baselines up to +1.5 BLEU score on five diverse domains in three language pairs and a real-world translation scenario.

Copied to clipboard

Shuo Huang, Xingliang Yuan, Gholamreza Haffari, Lizhen Qu

Domain Generalisation of NMT: Fusing Adapters with Leave-One-Domain-Out Training (2022.findings-acl)

Challenge:	Existing methods for text anonymization and de-identification struggle to balance privacy preservation with text naturalness and utility.
Approach:	They propose a tree-search-based iterative sentence rewriting algorithm that obfuscates or deletes private information while preserving coherence, relevance, and naturalness.
Outcome:	The proposed algorithm outperforms existing baselines on privacy-sensitive datasets.

Copied to clipboard

Thuy-Trang Vu, Shahram Khadivi, Dinh Phung, Gholamreza Haffari

Green & Sustainable NLP

Challenge:	Recent advances in neural machine translation (NMT) research have found NMT models sensitive to distribution shift and adversarial examples.
Approach:	They propose a leave-one-domain-out training strategy that learns to combine domain-specific parameters to avoid information leaking.
Outcome:	The proposed method outperforms baselines on three language pairs on average.

Harnessing Cross-lingual Features to Improve Cognate Detection for Low-resource Languages (2020.coling-main)

Copied to clipboard

Diptesh Kanojia, Raj Dabre, Shubham Dewangan, Pushpak Bhattacharyya, Gholamreza Haffari, Malhar Kulkarni

ACCESS : A Benchmark for Abstract Causal Event Discovery and Reasoning (2025.naacl-long)

Challenge:	a study of 14 Indian languages shows that cognates can be detected by word embeddings . cognates are variants of the same lexical form across languages .
Approach:	They propose to use cross-lingual word embeddings to detect cognates among 14 Indian languages . they then evaluate the impact of their method on neural machine translation .
Outcome:	The proposed method improves on a dataset of 12 Indian languages . it also improves quality of the extracted cognates by up to 2.76 BLEU .

Copied to clipboard

Vy Vo, Lizhen Qu, Tao Feng, Yuncheng Hua, Xiaoxi Kang, Songhai Fan, Tim Dwyer, Lay-Ki Soon, Gholamreza Haffari

FACTUAL: A Benchmark for Faithful and Consistent Textual Scene Graph Parsing (2023.findings-acl)

Challenge:	Existing methods for identifying event causality in NLP are limited in their scale and rely on lexical cues.
Approach:	They propose a benchmark for identifying abstract causality from a large-scale dataset.
Outcome:	The proposed benchmark can be leveraged for enhancing QA reasoning performance in LLMs.

Copied to clipboard

Zhuang Li, Yuyang Chai, Terry Yue Zhuo, Lizhen Qu, Gholamreza Haffari, Fei Li, Donghong Ji, Quan Hung Tran

Multimodality

Challenge:	Existing parsers that convert image captions into scene graphs often suffer from errors and inconsistency.
Approach:	They propose a dataset that re-annotates image captions using a new intermediate representation called FACTUAL-MR and a metric to measure scene graph similarity.
Outcome:	The proposed parser outperforms existing parsers in terms of faithfulness and consistency on multiple benchmark datasets.

Few-Shot Complex Knowledge Base Question Answering via Meta Reinforcement Learning (2020.emnlp-main)

Copied to clipboard

Yuncheng Hua, Yuan-Fang Li, Gholamreza Haffari, Guilin Qi, Tongtong Wu

Koala: An Index for Quantifying Overlaps with Pre-training Corpora (2023.emnlp-demo)

Challenge:	Existing approaches to complex question-answering (CQA) exhibit uneven performance when questions have different types, harboring inherently different characteristics, e.g., difficulty level.
Approach:	They propose a meta-reinforcement learning approach to program induction in CQA to tackle the potential distributional bias in questions.
Outcome:	The proposed method achieves state-of-the-art performance on the CQA dataset while using only five trial trajectories for the top-5 retrieved questions in each support set.

Copied to clipboard

Thuy-Trang Vu, Xuanli He, Gholamreza Haffari, Ehsan Shareghi

Reshaping Representation Space to Balance the Safety and Over-rejection in Large Audio Language Models (2025.emnlp-main)

Challenge:	Recent studies have shown that large language models can be influenced by the frequency of overlap between pre-training corpora.
Approach:	They propose to search over large pre-training corpora using lossless compressed suffix arrays with highly efficient compression rate and search support.
Outcome:	Koala is a searchable index over large pre-training corpora using lossless compressed suffix arrays with highly efficient compression rate and search support.

Copied to clipboard

Hao Yang, Lizhen Qu, Ehsan Shareghi, Gholamreza Haffari

Scene Graph Modification Based on Natural Language Commands (2020.findings-emnlp)

Challenge:	Large Audio Language Models (LALMs) have demonstrated unprecedented capabilities in natural language understanding and generation, revolutionizing human-machine dialogue.
Approach:	They propose an unsupervised safety-fine-tuning strategy that reshapes LALMs representation space to enhance existing LALM safety-alignment while balancing the risk of over-rejection.
Outcome:	The proposed approach improves LALMs safety under three input conditions while increasing over-rejection rate by only 0.88% on average.

Copied to clipboard

Xuanli He, Quan Hung Tran, Gholamreza Haffari, Walter Chang, Zhe Lin, Trung Bui, Franck Dernoncourt, Nhan Dam

Structured Data in NLP

Challenge:	Numerous parsing methods have been developed for a single sentence, while a typical human-computer interaction session or conversation is not singleturn.
Approach:	They propose to modify an existing scene graph given a new user's command by using graph-based sparse transformer and cross attention information fusion to improve performance.
Outcome:	The proposed models outperform previous systems adapted from the machine translation and graph generation literature and contribute to the research community.

IntroLM: Introspective Language Models via Prefilling-Time Self-Evaluation (2026.findings-acl)

Copied to clipboard

Hossein Hosseini Kasnavieh, Gholamreza Haffari, Christopher Leckie, Adel N. Toosi

Personal Information Leakage Detection in Conversations (2020.emnlp-main)

Challenge:	Existing methods to predict output quality of large language models rely on external classifiers with limited context windows and constrained representational capacity.
Approach:	They propose a method that enables causal language models to predict their own output quality during the prefilling phase without affecting generation using [CPX] tokens.
Outcome:	The proposed method outperforms existing classifiers on Qwen3-8B and DeBERTa-v3-Large models by 14% on question-answering benchmarks.

Copied to clipboard

Qiongkai Xu, Lizhen Qu, Zeyu Gao, Gholamreza Haffari

Learning How to Active Learn by Dreaming (P19-1)

Challenge:	4.5 billion dollars will be invested in conversational assistants (chatbots) by 2021, according to Opus Research 2 . Among diverse types of chatbots, Google Duplex represents the kind of AI personal assistants that act on behalf of people to perform simple tasks.
Approach:	They propose to protect personal information by warning users of detected suspicious sentences . they propose to use a constrained alignment problem to perform an alignment optimization problem .
Outcome:	The proposed models outperform baseline models on the behavior of personalized chit-chat dialogue systems.

Copied to clipboard

Thuy-Trang Vu, Ming Liu, Dinh Phung, Gholamreza Haffari

Cross-Lingual Transfer

Challenge:	Recent active learning methods are limited when the data distribution of learning problems vary.
Approach:	They propose a wake-and-dream-based active learning method that learns the AL policy directly on the target domain of interest by using wake and dream cycles.
Outcome:	The proposed method improves on cross-domain and cross-lingual tasks.

CosMo: Conditional Seq2Seq-based Mixture Model for Zero-Shot Commonsense Question Answering (2020.coling-main)

Copied to clipboard

Farhad Moghimifar, Lizhen Qu, Yue Zhuo, Mahsa Baktashmotlagh, Gholamreza Haffari

Learning to Explain: Generating Stable Explanations Fast (2021.acl-long)

Challenge:	Identifying the implicit causes and effects of a social context is the driving capability of commonsense reasoning.
Approach:	They propose a conditional seq2seq-based mixture model which generates context-dependent clauses for commonsense reasoning.
Outcome:	The proposed model improves on the current state-of-the-art models by +5.2% over existing models.

Copied to clipboard

Xuelin Situ, Ingrid Zukerman, Cecile Paris, Sameen Maruf, Gholamreza Haffari

Explainability & Interpretability in NLP

Challenge:	Existing methods for explaining outcome of machine learning models produce explanations, or rationales, which identify the attributions of features in an input example.
Approach:	They propose a Learning to Explain approach that learns the behaviour of an underlying explanation algorithm simultaneously from all training examples.
Outcome:	The proposed approach is 5 to 7.5104 times faster than existing models and has comparable faithfulness to the black-box model.

Continual Learning of Large Language Models (2025.emnlp-tutorials)

Copied to clipboard

Tongtong Wu, Trang Vu, Linhao Luo, Gholamreza Haffari

Understanding Unnatural Questions Improves Reasoning over Text (2020.coling-main)

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 .

Copied to clipboard

Xiaoyu Guo, Yuan-Fang Li, Gholamreza Haffari

Neural-Symbolic Commonsense Reasoner with Relation Predictors (2021.acl-short)

Challenge:	Complex question answering (CQA) requires large amounts of human-annotated data . learning effective CQA requires large amount of human annotated .
Approach:	They propose to map human-generated questions into unnatural machine-generated ones . they generate synthetic pairs and train a parser that associates synthetic questions with their corresponding action sequences.
Outcome:	The proposed model outperforms the state-of-the-art model trained on human-labeled data.

Copied to clipboard

Farhad Moghimifar, Lizhen Qu, Terry Yue Zhuo, Gholamreza Haffari, Mahsa Baktashmotlagh

Generate, Annotate, and Learn: NLP with Synthetic Text (2022.tacl-1)

Challenge:	Existing models for commonsense reasoning are limited by their limited set of facts, rendering them unfit for reasoning over new unseen situations and events.
Approach:	They propose a neural-symbolic reasoner which can combine commonsense facts with large-scale dynamic CKGs to draw conclusions about ordinary situations.
Outcome:	The proposed model outperforms the state-of-the-art models on the task of link prediction on CKGs.

Copied to clipboard

Xuanli He, Islam Nassar, Jamie Kiros, Gholamreza Haffari, Mohammad Norouzi

Self-supervised Rewiring of Pre-trained Speech Encoders: Towards Faster Fine-tuning with Less Labels in Speech Processing (2022.findings-emnlp)

Challenge:	Existing methods to generate unlabeled text are difficult to find.
Approach:	They propose a general framework called "generate, annotate, and learn" to take advantage of synthetic text within knowledge distillation, self-training, and few-shot learning applications.
Outcome:	The proposed framework achieves state-of-the-art knowledge distillation results for 6-layer transformers on the GLUE leaderboard.

Copied to clipboard

Hao Yang, Jinming Zhao, Gholamreza Haffari, Ehsan Shareghi

Audio Is the Achilles’ Heel: Red Teaming Audio Large Multimodal Models (2025.naacl-long)

Challenge:	Pre-trained speech encoders have facilitated great success across various speech processing tasks, but fine-tuning them for downstream tasks requires large training data to converge or to achieve state-of-the-art.
Approach:	They propose to rewire pre-trained speech encoders to improve their representation space without task-specific labels by neutrally synthesising audio inputs and frame masking.
Outcome:	The proposed model shows consistent improvement in isotropy in the representation space on 6 speech processing tasks.

Copied to clipboard

Hao Yang, Lizhen Qu, Ehsan Shareghi, Gholamreza Haffari

Context Dependent Semantic Parsing: A Survey (2020.coling-main)

Challenge:	Large Language Models (LMMs) have demonstrated ability to interact with humans through text . however, safety of audio LMMs remains under-explored .
Approach:	They red team the safety of five audio LMMs under three settings . they find that audio Lmms suffer an average attack success rate of 69.14% on harmful questions .
Outcome:	a new study shows that audio LMMs suffer an average success rate on harmful questions . the authors also show that the models exhibit safety vulnerabilities when distracted .

Copied to clipboard

Zhuang Li, Lizhen Qu, Gholamreza Haffari

Towards Inference-time Scaling for Continuous Space Reasoning (2026.findings-acl)

Challenge:	Semantic parsing is the task of translating natural language utterances into machine-readable meaning representations.
Approach:	They propose to use contextual information to translate natural language utterances into machine-readable meaning representations.
Outcome:	The proposed methods do not utilize contextual information, which could boost the semantic parsing systems.

Copied to clipboard

Minghan Wang, Thuy-Trang Vu, Ehsan Shareghi, Gholamreza Haffari

Few-Shot Semantic Parsing for New Predicates (2021.eacl-main)

Challenge:	Recent advances in reasoning large language models have expanded along training and inferencetime dimensions.
Approach:	They propose to use COCONUT (CITATION) continuous space reasoning LM as the backbone to generate diverse reasoning paths through dropout-based sampling.
Outcome:	The proposed method could enable performance gains similar to those observed in the discrete space, but only marginally improves in the continuous space.

Copied to clipboard

Zhuang Li, Lizhen Qu, Shuo Huang, Gholamreza Haffari

Importance-Aware Data Augmentation for Document-Level Neural Machine Translation (2024.eacl-long)

Challenge:	a recent study shows that state-of-the-art neural semantic parsers are less accurate when there is only a handful of utterance-logical form pairs per predicate.
Approach:	They propose to use a meta-learning method to train a few-shot learning problem . they also propose to regularize attention scores with alignment statistics and apply a smoothing technique .
Outcome:	The proposed method outperforms baselines in one and two-shot settings.

Copied to clipboard

Minghao Wu, Yufei Wang, George Foster, Lizhen Qu, Gholamreza Haffari

Effective Unsupervised Domain Adaptation with Adversarially Trained Language Models (2020.emnlp-main)

Challenge:	Document-level neural machine translation (DocNMT) models can be difficult and expensive to train due to data sparsity.
Approach:	They propose an Importance-Aware Data Augmentation algorithm that augments training data based on token importance information estimated by the norm of hidden states and training gradients.
Outcome:	The proposed algorithm outperforms strong DocNMT baselines and several data augmentation approaches on three widely-used benchmarks.

Copied to clipboard

Thuy-Trang Vu, Dinh Phung, Gholamreza Haffari

Monash University’s Submissions to the WNGT 2019 Document Translation Task (D19-56)

Challenge:	Recent work has shown the importance of training contextualised word embedding models on the domain of the target task of interest.
Approach:	They propose a masking strategy which adversarially masks out those tokens which are harder to reconstruct by the underlying MLM.
Outcome:	The proposed training strategy outperforms random masking on six unsupervised domain adaptation tasks and achieves up to +1.64 F1 score improvements.

Copied to clipboard

Sameen Maruf, Gholamreza Haffari

Less is More: Mitigate Spurious Correlations for Open-Domain Dialogue Response Generation Models by Causal Discovery (2023.tacl-1)

Challenge:	Despite the boom of work on document-level machine translation in the past two years, there has been a lack of the application of the proposed approaches to MT shared tasks.
Approach:	They propose to employ an established document-level neural machine translation model for the shared task of Rotowire document translation organised by the 3rd Workshop on Neural Generation and Translation (WNGT 2019).
Outcome:	The proposed model achieves a BLEU score of 39.83 for En-De and 45.06 for De-En translation directions on the Rotowire test set.

Copied to clipboard

Tao Feng, Lizhen Qu, Gholamreza Haffari

Contextual Neural Machine Translation Improves Translation of Cataphoric Pronouns (2020.acl-main)

Challenge:	Existing models suffer from spurious correlations and generate irrelevant and generic responses.
Approach:	They propose a model-agnostic method for training and inference using a conditional independence classifier that overcomes data sparsity.
Outcome:	The proposed method outperforms the baseline models in relevance, informativeness, and fluency.

Copied to clipboard

KayYen Wong, Sameen Maruf, Gholamreza Haffari

CultureInstruct: Curating Multi-Cultural Instructions at Scale (2025.naacl-long)

Challenge:	Recent studies have focused on past sentences as context with a focus on anaphora translation.
Approach:	They propose to use future context to improve NMT performance by comparing a contextual NMT model trained with past context to a context-agnostic model.
Outcome:	The proposed model outperforms the context-agnostic Transformer and shows comparable and in some cases improved performance.

Copied to clipboard

Viet Thanh Pham, Zhuang Li, Lizhen Qu, Gholamreza Haffari