Papers by Zhi Yang

15 papers
DevEval: A Manually-Annotated Code Generation Benchmark Aligned with Real-World Code Repositories (2024.findings-acl)

Copied to clipboard

Challenge: Existing benchmarks are poorly aligned with real-world code repositories and are insufficient to evaluate the coding abilities of Large Language Models (LLMs).
Approach: They propose a repository-level benchmark named DevEval to evaluate LLMs' coding abilities in real-world code repositories.
Outcome: The proposed benchmarks show that the LLMs perform better in real-world code repositories than existing benchmarks.
From Detection to Understanding: Multi-Turn Reasoning for Video Misinformation Analysis (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks focus on binary veracity judgments and do not evaluate process-level justifications for misinformation models.
Approach: They propose a video misinformation analysis benchmark that assesses reasoning in video misinterpretation.
Outcome: The proposed framework improves reasoning accuracy and explanation quality compared to existing models . it covers 12 fine-grained deception categories and progresses from perceptual attribution to intent and persuasion analysis.
BrailleLLM: Braille Instruction Tuning with Large Language Models for Braille Domain Tasks (2025.emnlp-main)

Copied to clipboard

Challenge: Existing Braille research focuses on isolated tasks while mixed-content Braille tasks face data scarcity and ambiguities.
Approach: They propose a syntax tree-based augmentation method tailored for Braille data.
Outcome: The proposed method improves Braille translation, formula-to-Braille conversion, and mixed-text translation.
TabularMath: Understanding Math Reasoning over Tables with Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Mathematical reasoning has long been a key benchmark for evaluating large language models.
Approach: They propose a framework that transforms math word problems into scalable tabular reasoning tasks.
Outcome: The proposed framework transforms math word problems into scalable and verified tabular reasoning tasks.
Guide the Many-to-One Assignment: Open Information Extraction via IoU-aware Optimal Transport (2023.acl-long)

Copied to clipboard

Challenge: Open Information Extraction (OIE) aims to extract structured information from text without the limitations of close ontology.
Approach: They propose a method to assign ground truth labels to parallelly generated tuple proposals . they leverage intersection-over-union (IoU) as assignment quality measurement .
Outcome: The proposed method outperforms the state-of-the-art models on three benchmarks.
On Safety Risks in Experience-Driven Self-Evolving Agents (2026.findings-acl)

Copied to clipboard

Challenge: Experience-driven self-evolution has emerged as a promising paradigm for improving the autonomy of large language model agents, yet its reliance on self-curated experience introduces underexplored safety risks.
Approach: They investigate how experience accumulation and utilization in self-evolving agents affect safety performance across web-based and embodied environments.
Outcome: The findings expose inherent limitations of current self-evolving agents and call for more principled strategies to ensure safe and reliable adaptation.
AutoEvolve: Automatically Evolving Queries for Applicable and Scalable Retrieval-Augmented Generation Benchmarking (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing automated generation methods exhibit Weak Applicability and Weak Scalability . existing methods are limited by their reliance on metadata from specific corpora .
Approach: They propose an approach to generate scalable RAG benchmarks using corpus-agnostic methods . they propose a difficulty-guided metric that directs query evolution process .
Outcome: The proposed approach evolves queries significantly more challenging than existing methods . it is able to dynamically increase difficulty, limiting scalability of the query .
Double-Checker: Large Language Model as a Checker for Few-shot Named Entity Recognition (2024.findings-emnlp)

Copied to clipboard

Challenge: Recent studies have demonstrated remarkable performance on few-shot Named Entity Recognition tasks due to the high cost of obtaining high-quality labeled data.
Approach: They propose to decompose the task into entity span detection and entity type classification using a type-independent entity span detector and then classify the detected spans based on their types.
Outcome: The proposed method consistently yields improvements over two baseline approaches.
VCSearch: Bridging the Gap Between Well-Defined and Ill-Defined Problems in Mathematical Reasoning (2025.emnlp-main)

Copied to clipboard

Challenge: Existing studies have improved the performance of Large language models on well-defined mathematical benchmarks, but they often overlook ill-defined problems.
Approach: They develop a large-scale benchmark that contains over 5,000 ill-defined mathematical problems.
Outcome: The proposed framework improves the accuracy of identifying unsolvable problems by at least 12% across different LLMs, thus achieving stronger robust mathematical reasoning ability.
Red Teaming Large Reasoning Models (2026.acl-long)

Copied to clipboard

Challenge: Large Reasoning Models (LRMs) have emerged as a powerful advancement in multi-step reasoning tasks, but they introduce safety and reliability risks, such as CoT-hijacking and prompt-induced inefficiencies.
Approach: They propose a unified benchmark to assess the trustworthiness of Large Reasoning Models.
Outcome: The proposed benchmark evaluates truthfulness, safety and efficiency on 26 models.
Generalization or Memorization: Data Contamination and Trustworthy Evaluation for Large Language Models (2024.findings-acl)

Copied to clipboard

Challenge: Considering the vast size and wide-ranging sources of LLMs’ training data, it could explicitly or implicitly include test data.
Approach: They propose a Contamination Detection via output Distribution (CDD) which detects data contamination only by identifying the peakedness of LLM's output distribution.
Outcome: The proposed method improves performance by 21.8%-30.2% on humanEval and TED: trustworthy evaluation via output distribution.
Basic Reading Distillation (2025.acl-long)

Copied to clipboard

Challenge: Large language models require high computational resources which limits their deployment in real-world applications.
Approach: They propose to distill large language models into smaller language models by either knowledge distillation or task distillation.
Outcome: The proposed model outperforms or performs comparable to over 20x bigger LLMs on language inference benchmarks and BIG-bench tasks.
LESA: Learnable LLM Layer Scaling-Up (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for depth scaling-up rely on empirical heuristic rules for layer duplication, resulting in poor initialization and slower convergence during continual pre-training.
Approach: They propose a method for learning latent parameters between layers by concatenating parameters from each layer and applying Singular Value Decomposition.
Outcome: Experiments show that LESA outperforms baseline models with less than half the cost of existing methods.
FinEval: A Chinese Financial Domain Knowledge Evaluation Benchmark for Large Language Models (2025.naacl-long)

Copied to clipboard

Challenge: Large language models have demonstrated outstanding performance in various natural language processing tasks, but their security capabilities in the financial domain have not been explored.
Approach: They propose to use a benchmark to evaluate large language models' financial domain knowledge and practical abilities.
Outcome: The proposed benchmark evaluates large language models' financial domain knowledge and practical abilities.
KuiLeiXi: a Chinese Open-Ended Text Adventure Game (2021.acl-demo)

Copied to clipboard

Challenge: Recent advances in pre-trained language models have made it possible to generate human-like text.
Approach: They propose to integrate an open-ended text adventure game in Chinese, named KuiLeiXi, where players interact with the AI until the plot goals are reached.
Outcome: The proposed game lacks incentives and relies on players to explore on their own.

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