Papers by Liyan Xu

13 papers
Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation (2021.emnlp-main)

Copied to clipboard

Challenge: Recent pre-trained language models have achieved remarkable zero-shot performance . we propose a self-learning framework that utilizes unlabeled data of target languages .
Approach: They propose a self-learning framework that utilizes unlabeled data of target languages to select silver labels for cross-lingual transfer tasks.
Outcome: The proposed framework outperforms baseline models on two cross-lingual tasks by 10 F1 on average and 2.5 accuracy on natural language inference (NLI).
ReFreeKV: Towards Threshold-Free KV Cache Compression (2026.findings-acl)

Copied to clipboard

Challenge: Towards the KV cache efficiency, we propose a new objective that lifts the threshold constraints for robust KV compression.
Approach: They propose a method that adjusts KV cache budgets while preserving full-cache performance.
Outcome: The proposed method can reduce memory consumption while preserving full-cache performance.
Identifying Factual Inconsistencies in Summaries: Grounding LLM Inference via Task Taxonomy (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing studies have focused on specialized BERT-variants and recent LLMs to reason inconsistencies.
Approach: They propose to incorporate task-specific taxonomy into inferences to facilitate both zero-shot and supervised paradigms.
Outcome: The proposed model outperforms specialized non-LLM and recent LLM models in a number of domains.
Understanding Factual Errors in Summarization: Errors, Summarizers, Datasets, Error Detectors (2023.acl-long)

Copied to clipboard

Challenge: Abstractive summarization systems still include factual errors in generated summaries despite recent improvements in factuality detection .
Approach: They aggregate factuality error annotations from nine existing datasets and stratify them according to the underlying summarization model.
Outcome: The proposed method improves on the ChatGPT-based model and shows that it is not superior for all error types.
Modeling Task Interactions in Document-Level Joint Entity and Relation Extraction (2022.naacl-main)

Copied to clipboard

Challenge: Existing work on document-level relation extraction has focused on end-to-end setting that extracts global entities and relations jointly.
Approach: They propose to introduce a two-way interaction between COREF and RE that is specifically designed to leverage task characteristics, bridging decisions of two tasks for direct task interference.
Outcome: The proposed model achieves the best performance by up to 2.3/5.1 F1 over the baseline.
ScEdit: Script-based Assessment of Knowledge Editing (2025.findings-acl)

Copied to clipboard

Challenge: Knowledge Editing (KE) has gained increasing attention, yet current evaluation frameworks do not integrate KE into real-world application scenarios.
Approach: They propose a script-based benchmark which encompasses both counterfactual and temporal edits and integrates token-level and text-level evaluation methods.
Outcome: The proposed method combines token-level and text-level evaluation methods with a new fact-based evaluation framework.
Revealing the Myth of Higher-Order Inference in Coreference Resolution (2020.emnlp-main)

Copied to clipboard

Challenge: Adapted coreference resolution models have only marginally improved performance over representation learning.
Approach: They implement an end-to-end coreference system and four HOI approaches to analyze the impact of higher-order inference on coreference resolution.
Outcome: The proposed model shows that the impact of higher-order inference (HOI) on coreference resolution is negative to marginal, providing a new perspective on the task.
Fine-Grained Modeling of Narrative Context: A Coherence Perspective via Retrospective Questions (2024.acl-long)

Copied to clipboard

Challenge: a novel graph for narrative comprehension captures coherence between passages in narratives . end-to-end paradigms are effective for comprehension tasks, but may not be sufficient for all comprehension scenarios.
Approach: They propose a graph dubbed NarCo which explicitly depicts task-agnostic coherence dependencies that are ready to be consumed by downstream tasks.
Outcome: The proposed graph is practically instantiated by LLMs without human annotations.
Exploring Deductive and Inductive Reasoning Capabilities of Large Language Models in Procedural Planning (2025.findings-emnlp)

Copied to clipboard

Challenge: Deductive and inductive reasoning are fundamental components of human cognition . authors present a benchmark to assess their performance in procedural planning .
Approach: They propose a benchmark to assess the deductive and inductive reasoning abilities of LLMs . they propose IMSE to enable LLM to generate multiple similar procedural plans .
Outcome: The proposed method improves inductive reasoning abilities of LLMs, the authors show . they show that LLM models show excellent deductive reasoning capabilities but suboptimal inductive performance.
Towards Open-World Product Attribute Mining: A Lightly-Supervised Approach (2023.acl-long)

Copied to clipboard

Challenge: Existing task setting for attribute mining on e-commerce products is closed-world, but recent work has moved towards open-world aspect.
Approach: They propose a task setting for attribute mining on e-commerce products that uses a high-quality seed attribute set bootstrapped from existing resources.
Outcome: The proposed approach surpasses baselines on existing attributes by 12 F1 and discovers values from 39% new attributes.
Dense Retrievers Can Fail on Simple Queries: Revealing The Granularity Dilemma of Embeddings (2025.findings-emnlp)

Copied to clipboard

Challenge: a limited number of text encoders are able to recognize fine-grained entities or events within encoded semantics.
Approach: They propose a new evaluation dataset to examine embeddings' ability to recognize fine-grained entities or events within encoded semantics.
Outcome: The proposed dataset shows embeddings struggle with fine-grained matching . the proposed encoder outperforms the state-of-the-art 7B model in a small sample .
Situated Embedding Models for Context-Aware Dense Retrieval (2026.acl-short)

Copied to clipboard

Challenge: Existing embedding models are not well-equipped to encode situated context effectively, i.e., situating a chunk’s meaning within its context.
Approach: They propose to represent short chunks in a way that is conditioned on a broader context window to enhance retrieval performance.
Outcome: The proposed model outperforms state-of-the-art embedding models on a book-plot retrieval dataset.
Online Coreference Resolution for Dialogue Processing: Improving Mention-Linking on Real-Time Conversations (2022.starsem-1)

Copied to clipboard

Challenge: Existing models for online decoding on active input are not trained to handle an online decode environment.
Approach: They propose a new direction of coreference resolution for online decoding on actively generated input such as dialogue . they propose to use models that accept utterances and their past context and find mentions upon each dialogue turn .
Outcome: The proposed model outperforms the baseline model by 10% on three datasets: Friends, OntoNotes, and BOLT.

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