Papers by Zhihao Yang

22 papers
Learning from Near-Misses: Error-Aware Contrastive Few-Shot Learning for NL2Formula (2026.acl-long)

Copied to clipboard

Challenge: Existing spreadsheet formulas often produce near-miss outputs due to an incorrect function, operator, or reference.
Approach: They propose an abstract syntax tree-based error taxonomy that organizes common error modes by the kind of decision that goes wrong in the parse tree.
Outcome: The proposed framework improves Exact Match (EM) by 6.4 points over supervised fine-tuning and matches self-consistency (SC@5) accuracy.
Exploring the Capability of Multimodal LLMs with Yonkoma Manga: The YManga Dataset and Its Challenging Tasks (2024.findings-emnlp)

Copied to clipboard

Challenge: YManga dataset is the first specifically designed for yonkoma manga understanding .
Approach: They propose to use a dataset of 1,015 yonkoma strips with 10,150 human annotations to define three tasks for panel sequence detection, intent generation and description generation for masked panels.
Outcome: The proposed dataset contains 1,015 high-quality yonkoma strips with 10,150 human annotations.
AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks evaluate agents in simplified, idealized settings, relying on pre-packaged tool interfaces, overlooking critical steps, and assume inputs are clean and fully specified.
Approach: They propose a framework that evaluates language agents in simplified, idealized settings . they show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .
Outcome: Experiments on 15 proprietary and open-source models show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .
OD-RTE: A One-Stage Object Detection Framework for Relational Triple Extraction (2023.acl-long)

Copied to clipboard

Challenge: Existing pipelines for relational triple extraction are underutilizing regional information of triple.
Approach: They propose a one-stage Object Detection framework for Relational Triple Extraction . framework uses vertices-based bounding box detection and global relational triple region detection .
Outcome: The proposed framework could extract all types of triples on two widely used datasets.
Exploiting Emotion-Semantic Correlations for Empathetic Response Generation (2023.findings-emnlp)

Copied to clipboard

Challenge: Empathetic response generation aims to generate empathetic responses by understanding the speaker’s emotional feelings from the language of dialogue.
Approach: They propose a dynamical Emotion-Semantic Correlation Model (ESCM) which constructs dynamic emotion-semantics through the interaction of context and emotions.
Outcome: The proposed model understands emotions more accurately and expresses fluent and informative empathetic responses.
Joint Entity and Relation Extraction for Legal Documents with Legal Feature Enhancement (2020.coling-main)

Copied to clipboard

Challenge: Existing methods for information extraction are based on pipelining to extract entities from unstructured judgment documents . a large number of judgment documents are released on China Judgments Online .
Approach: They propose a legal triplet extraction system for drug-related criminal judgment documents . they annotate a dataset for Named Entity Recognition and Relation Extraction in Chinese legal domain .
Outcome: The proposed system extracts entities and semantic relations jointly and benefits from the proposed legal lexicon feature and multi-task learning framework.
PFDial: A Structured Dialogue Instruction Fine-tuning Method Based on UML Flowcharts (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have shown remarkable progress in dialogue and reasoning, but they struggle to solve strictly constrained dialogue tasks.
Approach: They construct a dataset that contains 12,705 high-quality Chinese dialogue instructions from 440 flowcharts containing 5,055 process nodes.
Outcome: The proposed model outperforms GPT-4o models on backward transitions and outperformed GPT-42 models on the same dataset.
Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods for detecting pre-training data in large language models rely on superficial features like prediction confidence and loss, resulting in mediocre performance.
Approach: They propose a new algorithm to analyze neuron activation patterns between training and non-training data in large language models to improve their performance.
Outcome: The proposed algorithm outperforms existing methods across three benchmarks and multiple LLMs.
Prototype Tuning: A Meta-Learning Approach for Few-Shot Document-Level Relation Extraction with Large Language Models (2025.findings-naacl)

Copied to clipboard

Challenge: Few-Shot Document-Level Relation Extraction (FSDLRE) aims to develop models capable of generalizing to new categories with minimal support examples.
Approach: They propose a meta-training approach to train Large Language Models to improve their ICL capabilities . they construct simulated episodes using relation types that do not overlap with test corpus .
Outcome: Experimental results show that the proposed approach outperforms baseline models on few-shot tasks.
Breaking the Boundaries: A Unified Framework for Chinese Named Entity Recognition Across Text and Speech (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to Named Entity Recognition (NER) tasks are limited by the complexity of the data and the potential connections between tasks.
Approach: They propose a task to break the boundaries between different modal NER tasks by using a unified data format for inputs from different modalités.
Outcome: The proposed task breaks the boundaries between different modal NER tasks and is a unified implementation of them.
“Barking up the Right Tree”, a GAN-Based Pun Generation Model through Semantic Pruning (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods for generating humorous puns are limited and require a broad spectrum of commonsense and worldly skills.
Approach: They propose a GAN-based approach that employs semantic pruning and contrastive learning to generate humorous puns using a model that captures the semantic nuances of puns.
Outcome: The proposed model produces semantically coherent and humorous puns while ensuring both correctness and humor.
EMTIR-GRPO: Efficient Multi-Tool Augmented Large Language Models via Reinforcement Learning (2026.findings-acl)

Copied to clipboard

Challenge: Tool-integrated reasoning (TIR) enables large language models to invoke external tools for tasks beyond their internal capacity but often suffers from tool overuse.
Approach: They propose an algorithm that uses a composite reward to model tool costs and tool efficiency.
Outcome: The proposed algorithm models heterogeneous tool costs and encourages more cost-effective tool-use strategies.
DAGCN: Distance-based and Aspect-oriented Graph Convolutional Network for Aspect-based Sentiment Analysis (2024.findings-naacl)

Copied to clipboard

Challenge: Recent advances in sentiment analysis tend to interference from local factors such as irrelevant words and edges, hindering the precise identification of opinion words.
Approach: They propose a distance-based syntactic weight and Aspect-Fusion Attention to solve this problem.
Outcome: The proposed model outperforms state-of-the-art models on three public datasets and verify its effectiveness.
A Reinforcement Learning Framework for Natural Question Generation using Bi-discriminators (C18-1)

Copied to clipboard

Challenge: Existing research on visual question generation is focused on training models to fit the annotated data set that makes them indifferent from other language generation tasks.
Approach: They propose to use two discriminators to enhance the training of a visual question generator to ask natural questions about an image.
Outcome: The proposed model outperforms state-of-the-art models in terms of automatic and human evaluation metrics.
Learning to Detect Noisy Labels Using Model-Based Features (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to reduce label noise rely on heuristics and sample losses.
Approach: They propose a method that transfers the noise distribution to a clean set and trains a model to distinguish noisy labels from clean ones using model-based features.
Outcome: Empirically, the proposed approach improves over strong baselines on a wide range of tasks including text classification and speech recognition.
Two Languages Are Better than One: Bilingual Enhancement for Chinese Named Entity Recognition (2022.coling-1)

Copied to clipboard

Challenge: Existing studies focus on internal features of Chinese named entity recognition, but neglect other lingual modalities.
Approach: They propose a bilingual enhancement module for Chinese Named Entity Recognition . they integrate rich English information into Chinese representation and use it to learn the interaction between bilinguals and dependent information within Chinese.
Outcome: The proposed model can learn the interaction of bilinguals and dependent information within Chinese.
Transfer Learning in Biomedical Named Entity Recognition: An Evaluation of BERT in the PharmaCoNER task (D19-57)

Copied to clipboard

Challenge: Existing methods for natural language processing are labor-intensive and skill-dependent . Currently, most biomedical natural language tasks focus on English documents .
Approach: They introduce a BERT benchmark to facilitate the research of PharmaCoNER task . they evaluate two baselines based on Multilingual BERT and BioBERT on the corpus .
Outcome: The proposed task is based on multilingual BERT and BioBERT on the PharmaCoNER corpus.
ICA-Proto: Iterative Cross Alignment Prototypical Network for Incremental Few-Shot Relation Classification (2023.findings-eacl)

Copied to clipboard

Challenge: Existing methods for relation classification suffer from the scarcity of manually annotated data.
Approach: They propose a novel relation classification model that incorporates query representation into the encoding of novel prototypes and utilizes iteratively to achieve more interaction.
Outcome: The proposed model outperforms the state-of-the-art model on two benchmark datasets.
Code Reffix: A Benchmark for Reflection-Guided Code Repair with Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing benchmarks focus on the repair generation capability of LLMs, lacking fine-grained evaluation of reflection.
Approach: They propose a benchmark with oracle reflections and a dual-task protocol to decouple evaluation of reflection from repair.
Outcome: The proposed benchmarks show that underperforming reflection capabilities remain a bottleneck for code repair.
PAM: Enhancing General Alignment of Large Reasoning Models through Priority-Aware Metacognition (2026.acl-long)

Copied to clipboard

Challenge: Existing studies indicate that System-2 thinking alone does not transfer to the general alignment domain.
Approach: They propose to use priority-aware metacognition to help LRMs understand human preferences and monitor and regulate their thinking process.
Outcome: The proposed model improves general alignment performance by 10 points on helpfulness and harmless benchmarks.
SAFE-QAQ: End-to-End Slow-Thinking Audio-Text Fraud Detection via Reinforcement Learning (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for fraud detection rely on transcribed text, lacking acoustic cues . a proposed framework for audio-based slow-thinking fraud detection eliminates transcription errors .
Approach: They propose a framework for audio-based slow-thinking fraud detection that eliminates transcription errors and rewards slow-thought reasoning by capturing fine-grained audio details.
Outcome: The proposed method improves accuracy, inference efficiency, and real-time processing capabilities.
WECA: A WordNet-Encoded Collocation-Attention Network for Homographic Pun Recognition (D18-1)

Copied to clipboard

Challenge: Homographic puns have a long history in human writing, widely used in written and spoken literature, which intended as jokes.
Approach: They propose a WordNet-encoded model to settle polysemy of homographic puns and a word weighted model for recognizing them.
Outcome: The proposed model can distinguish between homographic pun and non-homographic pun texts.

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