Papers by Jiawei Cai

6 papers
Enhancing Language Representation with Constructional Information for Natural Language Understanding (2023.acl-long)

Copied to clipboard

Challenge: Recent advances in natural language processing focus on acquiring lexico-semantic information.
Approach: They propose a construction grammar which highlights the pairings of form and meaning to enrich language representation.
Outcome: The proposed model is superior to existing models on a variety of NLU tasks.
FCGEC: Fine-Grained Corpus for Chinese Grammatical Error Correction (2022.findings-emnlp)

Copied to clipboard

Challenge: grammatical error correction (GEC) is a complex task that requires high-quality data from native speakers.
Approach: They propose a human-annotated corpus to detect, identify and correct grammatical errors in Chinese examinations.
Outcome: The proposed model outperforms other models in low-resource settings, but there is a significant gap between the models and humans that encourages future models to bridge it.
KoCo: Conditioning Language Model Pre-training on Knowledge Coordinates (2026.acl-long)

Copied to clipboard

Challenge: Standard Large Language Model (LLM) pretraining treats corpora as flattened token sequences . a new method that maps every document into a three-dimensional semantic coordinate can bridge this gap .
Approach: They propose a method that maps every document into a three-dimensional semantic coordinate . they say it equips the model with explicit contextual awareness to learn the documents .
Outcome: Experiments show that knowledge coordinates help model distinguish stable facts from noise . authors say that the method significantly improves performance across 10 downstream tasks .
CMIG: Conceptual Metaphor Theory-Inspired Framework for Metaphorical Image Generation (2026.findings-acl)

Copied to clipboard

Challenge: Existing text-to-image systems often produce visually plausible but semantically literal outputs.
Approach: They propose a structured prompting framework inspired by Conceptual Metaphor Theory . they propose to identify source–target mappings, filter projectable source attributes and select a visual realization strategy in a reproducible reasoning workflow.
Outcome: The proposed framework improves semantic alignment and controllability on metaphor prompts.
CoELM: Construction-Enhanced Language Modeling (2024.acl-long)

Copied to clipboard

Challenge: Recent studies show that integrating constructional information can improve the performance of pre-trained language models.
Approach: They propose a construction-Enhanced language model that embeds constructional semantics into language models for natural language generation.
Outcome: The proposed model outperforms existing models on various benchmarks.
SampleMix: A Sample-wise Pre-training Data Mixing Strategy by Coordinating Data Quality and Diversity (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for pretraining data mixing for large language models neglect significant inter-domain overlaps and commonalities, failing to control the global diversity of the constructed training dataset.
Approach: They propose a sample-wise data mixture approach that performs global cross-domain sampling by systematically evaluating the quality and diversity of each sample.
Outcome: The proposed method exceeds existing domain-based methods in multiple downstream tasks and perplexity assessments.

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