Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Claire CardieSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works54 from public data
- Learning to Ask: Neural Question Generation for Reading Comprehension763
An attention-based sequence learning model for the task and the effect of encoding sentence- vs. paragraph-level information is investigated and results show that the system significantly outperforms the state-of-the-art rule-based system.
- Event Extraction by Answering (Almost) Natural Questions505
This work introduces a new paradigm for event extraction by formulating it as a question answering (QA) task, which extracts the event arguments in an end-to-end manner and outperforms prior methods substantially.
- Harvesting Paragraph-level Question-Answer Pairs from Wikipedia178
It is found that the linguistic knowledge introduced by the coreference representation aids question generation significantly, producing models that outperform the current state-of-the-art.
- PRD: Peer Rank and Discussion Improve Large Language Model based Evaluations146
The peer rank (PR) algorithm that takes into account each peer LLM's pairwise preferences of all answer pairs, and outputs a final ranking of models; and peer discussion (PD), where two LLMs are prompt to discuss and try to reach a mutual agreement on the preferences of two answers, are proposed.
- Document-Level Event Role Filler Extraction using Multi-Granularity Contextualized Encoding119
This work proposes a novel multi-granularity reader to dynamically aggregate information captured by neural representations learned at different levels of granularity, and evaluates the models on the MUC-4 event extraction dataset, showing that the best system performs substantially better than prior work.
- LLM4SR: A Survey on Large Language Models for Scientific Research106
This paper presents the first systematic survey dedicated to exploring how LLMs are revolutionizing the scientific research process, and analyzes the unique roles LLMs play across four critical stages of research: hypothesis discovery, experiment planning and implementation, scientific writing, and peer reviewing.
- Identifying Where to Focus in Reading Comprehension for Neural Question Generation102
A hierarchical neural sentence-level sequence tagging model that is fully data-driven — with no sophisticated NLP pipelines or any hand-crafted rules/features — and compares favorably to a number of baselines when evaluated on the SQuAD data set.
- GRIT: Generative Role-filler Transformers for Document-level Event Entity Extraction91
A generative transformer-based encoder-decoder framework (GRIT) that is designed to model context at the document level: it can make extraction decisions across sentence boundaries; is implicitly aware of noun phrase coreference structure, and has the capacity to respect cross-role dependencies in the template structure.
- MLR-Copilot: Autonomous Machine Learning Research based on Large Language Models Agents71
MLR-COPILOT, an autonomous Machine Learning Research framework powered by large language model agents, is presented, designed to enhance ML research productivity through automatic generation and implementation of research ideas within constraints.
- Few-shot Intent Classification and Slot Filling with Retrieved Examples62
This paper proposes a span-level retrieval method that learns similar contextualized representations for spans with the same label via a novel batch-softmax objective and uses the labels of the retrieved spans to construct the final structure with the highest aggregated score.
- FaithScore: Fine-grained Evaluations of Hallucinations in Large Vision-Language Models55
A reference-free and fine-grained evaluation metric that measures the faithfulness of the generated free-form answers from large vision-language models (LVLMs), which reveals that current systems are prone to generate hallucinated content unfaithful to the image, which leaves room for future improvements.
- Dynamic Global Memory for Document-level Argument Extraction48
A new global neural generation-based framework for document-level event argument extraction is introduced by constructing a document memory store to record the contextual event information and leveraging it to implicitly and explicitly help with decoding of arguments for later events.
- Template Filling with Generative Transformers45
Experiments demonstrate that this framework substantially outperforms pipeline-based approaches, and other neural end-to-end baselines that do not model between-event dependencies, and improves performance on documents containing multiple events.
- Logical Reasoning over Natural Language as Knowledge Representation: A Survey41
This paper provides a comprehensive overview on a new paradigm of logical reasoning, which uses natural language as knowledge representation and pretrained language models as reasoners, including philosophical definition and categorization of logical reasoning.
- FGAIF: Aligning Large Vision-Language Models with Fine-grained AI Feedback39
This work proposes an innovative method to align modalities in LVLMs through Fine-Grained Artificial Intelligence Feedback (FGAIF), which mainly consists of three steps: AI-based Feedback Collection, Fine-grained Reward Model Training, and Reinforcement Learning with Fine-grained Reward.
- RESIN-11: Schema-guided Event Prediction for 11 Newsworthy Scenarios39
This work introduces RESIN-11, a new schema-guided event extraction&prediction framework that can be applied to a large variety of newsworthy scenarios and builds a demo website based on the dockerized system and schema library.
- QA-Driven Zero-shot Slot Filling with Weak Supervision Pretraining34
This work introduces QA-driven slot filling (QASF), which extracts slot-filler spans from utterances with a span-based QA model, and uses a linguistically motivated questioning strategy to turn descriptions into questions, allowing the model to generalize to unseen slot types.
- MEQA: A Benchmark for Multi-hop Event-centric Question Answering with Explanations33
A novel semi-automatic question generation strategy by composing event structures from information extraction datasets and the first Multi-hop Event-centric Question Answering (MEQA) benchmark is presented, which shows that MEQA is challenging for the latest state-of-the-art models encompassing large language models (LLMs); and how they fall short of providing faithful explanations of the event-centric reasoning process.
- 33
- Be Consistent! Improving Procedural Text Comprehension using Label Consistency31
This work presents a new learning framework that leverages label consistency during training, allowing consistency bias to be built into the model and significantly improves prediction performance over prior state-of-the-art systems.
- 26
- Improving Event Duration Prediction via Time-aware Pre-training24
This work introduces two effective models for duration prediction, which incorporate external knowledge by reading temporal-related news sentences (time-aware pre-training), and demonstrates these models are capable of duration prediction in the unsupervised setting, outperforming the baselines.
- Logical Entity Representation in Knowledge-Graphs for Differentiable Rule Learning23
Empirical results demonstrate that the proposed Logical Entity RePresentation (LERP) model outperforms other rule learning methods in knowledge graph completion and is comparable or even superior to state-of-the-art black-box methods.
- 22
- LDC: Learning to Generate Research Idea with Dynamic Control21
A novel framework that employs a two-stage approach combining Supervised Fine-Tuning and controllable Reinforcement Learning to provide a balanced approach to research ideation, achieving high-quality outcomes by dynamically navigating the trade-offs among novelty, feasibility, and effectiveness.
- End-to-end Case-Based Reasoning for Commonsense Knowledge Base Completion20
This framework addresses a fundamental question on whether CBR methodology can be utilized to improve deep learning models and generates more valid, informative, and novel knowledge than the state-of-the-art COMET model for CKBC in both the fully supervised and few-shot settings.
- Dr.V : A Hierarchical Perception-Temporal-Cognition Framework to Diagnose Video Hallucination by Fine-Grained Spatial-Temporal Grounding16
Dr.V, a hierarchical framework covering perceptive, temporal, and cognitive levels to diagnose and mitigate video hallucination by fine-grained spatial-temporal grounding, is proposed, offering a practical blueprint for robust video understanding in real-world scenarios.
- 16
- HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation15
Hierarchical Process Rewards for Efficient agentic RAG (HiPRAG), a training methodology that incorporates a fine-grained, knowledge-grounded process reward into the RL training, demonstrates the importance and potential of fine-grained control through RL, for improving the efficiency and optimality of reasoning for search agents.
- FG-PRM: Fine-grained Hallucination Detection and Mitigation in Language Model Mathematical Reasoning15
The FG-PRM (Fine-Grained Process Reward Model), an augmented model designed to detect and mitigate hallucinations in a fine-grained, step-level manner, excels in fine-grained hallucination detection and substantially boosts the performance of LLMs on GSM8K and MATH benchmarks.
- Automatic Error Analysis for Document-level Information Extraction13
This work builds on the work of Kummerfeld and Klein (2013) to propose a transformation-based framework for automating error analysis in document-level event and (N-ary) relation extraction, and compares two state-of-the-art document- level template-filling approaches on datasets from three domains.
- Document-level Event-based Extraction Using Generative Template-filling Transformers.13
A generative transformer-based encoder-decoder framework that is designed to model context at the document level: it can make extraction decisions across sentence boundaries; is implicitly aware of noun phrase coreference structure, and has the capacity to respect cross-role dependencies in the template structure.
- CLUE: Non-parametric Verification from Experience via Hidden-State Clustering12
Clue (Clustering and Experience-based Verification), a deliberately minimalist, non-parametric verifier that consistently outperforms LLM-as-a-judge baselines and matches or exceeds modern confidence-based methods in reranking candidates, improving both top-1 and majority-vote accuracy across AIME 24/25 and GPQA.
- 12
- 11
- FIHA: Autonomous Hallucination Evaluation in Vision-Language Models with Davidson Scene Graphs9
The FIHA (autonomous Fine-graIned Hallucination evAluation evaluation in LVLMs) is introduced, which could access hallucination LVLMs in the LLM-free and annotation-free way and model the dependency between different types of hallucinations.
- 8
- A Comprehensive Analysis for Visual Object Hallucination in Large Vision-Language Models6
This paper analyzes each component of LLaVA-like LVLMs -- the large language model, the vision backbone, and the projector -- to identify potential sources of error and their impact, and proposes methods to mitigate hallucination for each problematic component.
- 6
- Toward Consistent and Informative Event-Event Temporal Relation Extraction5
This model combines a pretrained language model and a graph neural network to output event embeddings, which captures the contextual information of event graphs and achieves state-of-the-art performance and generates high-quality event graphs.
- Cornell Belief and Sentiment System at TAC 2016.5
The 2016 system of the CornPittMich team for the TAC Belief and Sentimen (BeSt) task for English and Chinese is described, employing a combination of simple, hand-crafted rules and machine learning-based approaches.
- 4
- 4
- Leveraging Structured Metadata for Improving Question Answering on the Web3
A neural passage selection model is proposed that leverages metadata information with a fine-grained encoding strategy, which learns the representation for metadata predicates in a hierarchical way, which significantly outperform baseline models, which do not incorporate metadata.
- Efficient Document-level Event Relation Extraction2
An innovative two-stage framework is proposed to tackle the challenges of ERE, consisting of a retriever to identify the related event pairs and a cross-encoder to classify the relationships between the retrieved pairs.
- 2
- Multimodal Reference Visual Grounding2
This work introduces a new task named Multimodal Reference Visual Grounding (MRVG), and introduces a novel method, named MRVG-Net, to solve this visual grounding problem, by efficiently using reference images with few-shot object detection and using Large Language Models (LLMs) for object matching.
- Do Retrieval-Augmented Language Models Adapt to Varying User Needs?2
This paper introduces a novel evaluation framework that systematically assesses RALMs under three user need cases-Context-Exclusive, Context-First, and Memory-First-across three distinct context settings: Context Matching, Knowledge Conflict, and Information Irrelevant.
- Making Natural Language Reasoning Explainable and Faithful2
This talk will focus on the design of leveraging structured information (that is grounded to the context), for the explainable complex question answering and reasoning, and the multi-module interpretable framework for inductive reasoning, which conducts step-wise faithful reasoning with iterative feedback.
- 2
- 2
- –
- –
- AGent: A Novel Pipeline for Automatically Creating Unanswerable Questions–
The usefulness of AGent, a novel pipeline that automatically creates new unanswerable questions by re-matching a question with a context that lacks the necessary information for a correct answer, is demonstrated.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.