Papers by Xinya Du

32 papers
Language Models as Inductive Reasoners (2024.eacl-long)

Copied to clipboard

Challenge: Inductive reasoning is a core component of human intelligence.
Approach: They propose a task to induce natural language rules from natural language facts using natural language as representation for knowledge instead of formal language.
Outcome: The proposed task surpasses baselines in both automatic and human evaluations.
Zero-Shot Classification by Logical Reasoning on Natural Language Explanations (2023.findings-acl)

Copied to clipboard

Challenge: Experimental results show that CLORE is superior to baselines on zero-shot classification tasks.
Approach: They propose a framework for classification by logically parsing and reasoning on natural language explanations.
Outcome: The proposed framework outperforms baselines on zero-shot classification tasks.
FaithScore: Fine-grained Evaluations of Hallucinations in Large Vision-Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) lack the capacity to handle multimodal inputs effectively.
Approach: They introduce a reference-free and fine-grained evaluation metric that measures the faithfulness of the generated free-form answers from large vision-language models.
Outcome: The proposed metric measures the faithfulness of free-form answers from large vision-language models.
Large Language Models for Automated Open-domain Scientific Hypotheses Discovery (2024.findings-acl)

Copied to clipboard

Challenge: Existing research on hypothetical induction is limited by the observation annotations in the dataset and the ground truth hypotheses are mostly commonsense knowledge.
Approach: They propose a first dataset for social science academic hypotheses discovery using raw web corpus as observations and propose valid, useful scientific hypothese . they propose 'a multi-module framework' that includes feedback mechanisms to boost performance.
Outcome: The proposed dataset generates valid, novel, and helpful scientific hypotheses, even new to humanity, using open-domain data and a web corpus as observations.
Retrieval-Augmented Generative Question Answering for Event Argument Extraction (2022.emnlp-main)

Copied to clipboard

Challenge: Existing methods to extract arguments from documents are based on generating and post-processing a complex target sequence (template).
Approach: They propose a retrieval-augmented generative QA model that retrieves the most similar QA pair and augments it as prompt to the current example's context, then decodes the arguments as answers.
Outcome: The proposed model outperforms prior methods across fully supervised, domain transfer, and fewshot learning settings and compares with clustering-based sampling strategies.
Document-Level Event Role Filler Extraction using Multi-Granularity Contextualized Encoding (2020.acl-main)

Copied to clipboard

Challenge: Document-level event extraction requires a view of a larger context to determine which spans of text correspond to event role fillers.
Approach: They propose a multi-granularity reader to dynamically aggregate information captured by neural representations learned at different levels of granularities.
Outcome: The proposed model performs substantially better than previous models on the MUC-4 event extraction dataset.
Automatic Error Analysis for Document-level Information Extraction (2022.acl-long)

Copied to clipboard

Challenge: Document-level information extraction (IE) tasks have been revisited in earnest . evaluation of the approaches has been limited in a number of dimensions .
Approach: They propose a transformation-based framework for automating error analysis in document-level event and (N-ary) relation extraction.
Outcome: The proposed framework compares two state-of-the-art document-level template-filling approaches on datasets from three domains and four systems from the MUC-4 evaluation.
Event Extraction by Answering (Almost) Natural Questions (2020.emnlp-main)

Copied to clipboard

Challenge: Existing work in event argument extraction relies heavily on entity recognition as a preprocessing/concurrent step, causing error propagation.
Approach: They propose a question answering task that extracts event arguments in an end-to-end manner.
Outcome: The proposed framework outperforms prior work on the ACE 2005 task on event argument extraction.
Leveraging Structured Metadata for Improving Question Answering on the Web (2020.aacl-main)

Copied to clipboard

Challenge: Using metadata information from web pages can improve the performance of answer passage selection/reranking models.
Approach: They propose a neural passage selection model that leverages metadata information with a fine-grained encoding strategy to learn the representation for metadata predicates in a hierarchical way.
Outcome: The proposed model outperforms baseline models on the MS MARCO and Recipe-MARCO datasets and shows that it is more accurate than baseline models.
Is Grokking Worthwhile? Functional Analysis and Transferability of Generalization Circuits in Transformers (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models struggle with the "curse of two-hop reasoning" in compositional tasks.
Approach: They propose to form a "Generalization Circuit" during a prolonged "grokking" phase . they argue that grokkking is the process of integrating memorized atomic facts into an easy-acquire reasoning path.
Outcome: The proposed model is superior to non-grokked models, but it requires a large computational cost . the study shows that grokking is not the sudden acquisition of a new reasoning paradigm .
Your Reasoning Model is Secretly a Reward Model - Optimization-Free Verification from Experience (2026.acl-long)

Copied to clipboard

Challenge: Existing verifiers operate on the surface text or on confidence proxies derived from token probabilities, which can be brittle.
Approach: They propose a training-free, non-parametric verifier that summarizes each reasoning trace by an activation delta and compares it to two class centroids computed from labeled experience.
Outcome: The proposed model improves selection and reranking on large and less-calibrated models.
SciEvent: Benchmarking Multi-domain Scientific Event Extraction (2025.emnlp-main)

Copied to clipboard

Challenge: Existing work on scientific information extraction relies on entity-relation extraction in narrow domains . current models struggle in domains such as sociology and humanities .
Approach: They propose a multi-domain benchmark for scientific abstract annotations using a unified event extraction schema.
Outcome: The proposed benchmark includes 500 abstracts across five research domains with manual annotations of event segments, triggers, and fine-grained arguments.
GRIT: Generative Role-filler Transformers for Document-level Event Entity Extraction (2021.eacl-main)

Copied to clipboard

Challenge: Document-level role-filler entity extraction (REE) is a classic problem in information extraction (IE) and NLP.
Approach: They propose a document-level role-filler entity extraction framework that can make extraction decisions across sentence boundaries and implicitly recognizes noun phrase coreference structure.
Outcome: The proposed framework performs better than previous models on a MUC-4 dataset.
FIFA: Unified Faithfulness Evaluation Framework for Text-to-Video and Video-to-Text Generation (2026.findings-acl)

Copied to clipboard

Challenge: Existing evaluation methods for VideoMLLMs are limited to one task and fail to assess hallucinations in open-ended, free-form responses.
Approach: They propose a unified framework that extracts comprehensive descriptive facts and models their semantic dependencies via a Spatio-Temporal Semantic Dependency Graph.
Outcome: The proposed framework aligns more closely with human judgment than existing evaluation methods and improves factual consistency in both text and video generation.
Harvesting Paragraph-level Question-Answer Pairs from Wikipedia (P18-1)

Copied to clipboard

Challenge: Existing models that only take into account sentence-level information do not generate question-answer pairs.
Approach: They propose a neural network approach that incorporates coreference knowledge via a novel gating mechanism for paragraphlevel question generation.
Outcome: The proposed model outperforms existing models on a Wikipedia article question-answer generation task.
Be Consistent! Improving Procedural Text Comprehension using Label Consistency (N19-1)

Copied to clipboard

Challenge: Existing systems for procedural text comprehension still struggle with this task . evaluative work shows that consistent predictions from multiple entities can improve performance .
Approach: They propose a framework that leverages label consistency during training to improve prediction performance.
Outcome: The proposed framework significantly improves prediction performance over previous state-of-the-art systems on a standard benchmark dataset for procedural text, ProPara.
Template Filling with Generative Transformers (2021.naacl-main)

Copied to clipboard

Challenge: Template filling tasks are usually tackled by a pipeline of two separate systems, one for role-filler extraction and another for template/event recognition.
Approach: They propose a framework that naturally models the dependence between entities within a single event and across multiple events described in a document.
Outcome: The proposed framework outperforms pipeline-based approaches and other neural baselines that do not model between-event dependencies on documents containing multiple events.
Leveraging Structured Information for Explainable Multi-hop Question Answering and Reasoning (2023.findings-emnlp)

Copied to clipboard

Challenge: Neural models, including large language models (LLMs), achieve superior performance on multi-hop question-answering tasks.
Approach: They propose to use the chain-of-thought mechanism to generate both the reasoning chain and the answer.
Outcome: Empirical results show that the proposed framework generates more faithful reasoning chains and significantly improves the QA performance on two benchmark datasets.
Improving Event Duration Prediction via Time-aware Pre-training (2020.findings-emnlp)

Copied to clipboard

Challenge: Understanding duration of event expressed in text is crucial task in NLP . current methods focus on developing features and cannot utilize external textual knowledge.
Approach: They propose two models that incorporate external knowledge by reading temporal-related news sentences.
Outcome: The proposed models outperform baseline models and capture duration information more accurately.
Dynamic Global Memory for Document-level Argument Extraction (2022.acl-long)

Copied to clipboard

Challenge: Recent work on document-level event argument extraction is restricted by sequence length constraints and ignores global context between events.
Approach: They propose to construct a document memory store to extract contextual event information and leverage it to implicitly and explicitly help with decoding of arguments for later events.
Outcome: The proposed framework outperforms prior methods and is more robust to adversarially annotated examples with constrained decoding design.
Probing Representations for Document-level Event Extraction (2023.findings-emnlp)

Copied to clipboard

Challenge: Document-level information extraction tasks require a more comprehensive understanding that often extends to the entire input document.
Approach: They propose to use probing to analyze document-level information extraction representations by embedding probes into a standard dataset.
Outcome: The proposed models improve argument detections but struggle with document length and cross-sentence discourse.
QAEVENT: Event Extraction as Question-Answer Pairs Generation (2024.findings-eacl)

Copied to clipboard

Challenge: a novel approach to document-level event extraction requires linguistic expertise.
Approach: They propose a question-and-answer-pair representation of document-level events as question and answer pairs (QAEVENT) this allows for more scalable and faster annotations from crowdworkers without linguistic expertise.
Outcome: The proposed method produces high-quality data for document-level event extraction compared to pre-defined schemas . it is scalable and faster than existing methods, with no linguistic expertise.
POE: Process of Elimination for Multiple Choice Reasoning (2023.emnlp-main)

Copied to clipboard

Challenge: Current language models perform well on multiple choice reasoning tasks, but the options are not treated equally.
Approach: They propose a two-step scoring method that scores options and masks them to make the final prediction from the remaining options.
Outcome: The proposed method is especially performant on logical reasoning tasks.
FG-PRM: Fine-grained Hallucination Detection and Mitigation in Language Model Mathematical Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods to detect hallucinations in large language models lack nuanced understanding of their types and manifestations.
Approach: They propose a taxonomy that categorizes hallucinations into six types . they propose an augmented model to detect and mitigate hallucinosity in a fine-grained manner .
Outcome: The proposed model detects and mitigates hallucinations in a fine-grained manner . it significantly boosts the performance of LLMs on GSM8K and MATH benchmarks.
LMR-BENCH: Evaluating LLM Agent’s Ability on Reproducing Language Modeling Research (2025.emnlp-main)

Copied to clipboard

Challenge: Large language model (LLM) agents have demonstrated remarkable potential in advancing scientific discovery, but their capability in reproducing code from research papers remains underexplored.
Approach: They propose to evaluate LLM agents' ability to reproduce scientific research papers by analyzing code reproduction tasks from 23 research papers published in top-tier NLP venues.
Outcome: The proposed benchmark systematically evaluates the capability of large language model (LLM) agents on code reproduction from Language Modeling Research.
End-to-end Case-Based Reasoning for Commonsense Knowledge Base Completion (2023.eacl-main)

Copied to clipboard

Challenge: Pretrained language models have been shown to store knowledge in their parameters and have achieved reasonable performance in knowledge-intensive tasks.
Approach: They propose to provide retrieved passages that contain relevant knowledge as additional input to the commonsense knowledge base completion (CKBC) task.
Outcome: The proposed framework generates more valid, informative, and novel knowledge than the state-of-the-art COMET model for commonsense knowledge base completion (CKBC) tasks.
Document-level Causal Relation Extraction with Knowledge-guided Binary Question Answering (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing research on Event-Event Causal Relation Extraction (ECRE) has highlighted the lack of document-level modeling and causal hallucinations.
Approach: They propose a Knowledge-guided binary Question Answering method with event structures for ECRE that utilizes cross-task knowledge in IE.
Outcome: The proposed method achieves state-of-the-art on the MECI and MAVEN-ERE datasets.
Search Wisely: Mitigating Sub-optimal Agentic Searches By Reducing Uncertainty (2025.emnlp-main)

Copied to clipboard

Challenge: Recent advances in Large Language Models (LLMs) have propelled their use in informationintensive tasks such as question answering and knowledge synthesis.
Approach: They propose a reinforcement learning-based training method that incorporates confidence thresholds to reward high-certainty search decisions.
Outcome: The proposed method outperforms baseline models on seven QA benchmarks and demonstrates that it is more efficient than existing methods.
Few-shot Intent Classification and Slot Filling with Retrieved Examples (2021.naacl-main)

Copied to clipboard

Challenge: Existing methods for few-shot learning are based on labeled examples, but they are non-trivial . few-sshot learning is challenging due to the imbalance in the amount of data between the source and target domains.
Approach: They propose retrieval-based methods for intent classification and slot filling tasks . they use a batch-softmax objective to learn similar contextualized representations for spans .
Outcome: The proposed method outperforms previous systems on the CLINC and SNIPS benchmarks.
FIHA: Automated Fine-grained Hallucinations Evaluations in Large Vision Language Models with Davidson Scene Graphs (2025.findings-acl)

Copied to clipboard

Challenge: Current approaches to large vision-language models rely on costly annotations and are not comprehensive in terms of evaluating all aspects.
Approach: They propose an automated method which can access LVLMs hallucination in an LLM-free and annotation-free way and model the dependency between different types of halluciNations.
Outcome: The proposed model can model the dependency between different types of hallucinations and generate Q&A pairs on any image dataset at minimal cost.
QA-Driven Zero-shot Slot Filling with Weak Supervision Pretraining (2021.acl-short)

Copied to clipboard

Challenge: Existing methods to predict slots and their values do not encode enough semantic information, limiting the models’ zero-shot capability.
Approach: They propose a QA-driven slot filling model which extracts slot-filler spans from utterances with a span-based QA model.
Outcome: The proposed model outperforms baselines by over 5% on the SNIPS benchmark.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations