Papers by Zhengyuan Wang

24 papers
Automatic Table Union Search with Tabular Representation Learning (2023.findings-acl)

Copied to clipboard

Challenge: Existing methods to identify uniability based on column representations are insufficient to reveal latent relational features to describe column relation between pair of columns.
Approach: They propose a self-supervised table union search framework called AutoTUS to learn column relational representations in a multi-stage manner.
Outcome: The proposed framework improves on the SOTA baseline and on real-world datasets.
In2Core: Leveraging Influence Functions for Coreset Selection in Instruction Finetuning of Large Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) exhibit surprising abilities across a variety of language tasks.
Approach: They propose an algorithm which selects a coreset by analyzing correlation between training and evaluation samples with a trained model.
Outcome: The proposed algorithm can achieve similar performance with just 50% of the training data while preserving the accuracy of the existing model.
SLOT: Structuring the Output of Large Language Models (2025.emnlp-industry)

Copied to clipboard

Challenge: Structured outputs are essential for large language models (LLMs) but often deviate from predefined schemas hampering reliable application development.
Approach: They propose a model-agnostic approach that transforms unstructured LLM outputs into precise structured formats.
Outcome: The proposed model-agnostic approach transforms unstructured LLM outputs into precise structured formats.
AudioBench: A Universal Benchmark for Audio Large Language Models (2025.naacl-long)

Copied to clipboard

Challenge: Existing evaluation regimes for audio large language models do not cover the breadth of their possible use cases.
Approach: They propose to use AudioBench to evaluate audio large language models . they found that no single model excels consistently across all tasks .
Outcome: The proposed evaluation targets speech understanding, audio scene understanding, and voice understanding (paralinguistic) . no single model excels consistently across all tasks, the paper found .
A Systematic Survey of Automatic Prompt Optimization Techniques (2025.emnlp-main)

Copied to clipboard

Challenge: Recent advances in prompt engineering have created impediments for end users to adopt . however, prompt engineering remains an impedance due to rapid advances in models, tasks, and associated best practices.
Approach: They propose to define APO as a 5-part unifying framework and categorize all relevant works based on their salient features.
Outcome: The proposed framework aims to improve the performance of large language models on various tasks.
Resilience of Large Language Models for Noisy Instructions (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) are powerful tools for interpreting human commands and generating text.
Approach: They examine the resilience of large language models against five common types of disruptions including ASR, OCR, grammatical errors, typographical errors and distractive content.
Outcome: The models show resistance to noise, but their performance suffers . authors evaluated the models against five common types of disruptions based on their results .
NameGuess: Column Name Expansion for Tabular Data (2023.emnlp-main)

Copied to clipboard

Challenge: Tabular data is used for storing and organizing information in web and enterprise applications.
Approach: They propose a task to expand column names as a natural language generation problem by conditioning on table content and column header names to improve auto-regressive models.
Outcome: The proposed task improves auto-regressive models on table content and column header names to match human performance.
Reinforcing Compositional Retrieval: Retrieving Step-by-Step for Composing Informative Contexts (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated remarkable capabilities across numerous tasks, yet they often rely on external context to handle complex tasks.
Approach: They propose a tri-encoder sequential retriever that models a Markov Decision Process (MDP) this method decomposes the probability of retrieving a set of elements into a sequence of conditional probabilities and allows each retrieval step to be conditioned on previously selected examples.
Outcome: The proposed method outperforms baselines and shows that it can handle multiple pieces of evidence or examples.
BoundRL: Efficient Token-level Structured Text Segmentation through Reinforced Boundary Generation (2026.findings-acl)

Copied to clipboard

Challenge: Structured texts often contain elements beyond plain language, such as code snippets, which conventional sentence-level segmentation methods cannot handle effectively.
Approach: They propose a token-level approach that performs efficient token-based text segmentation and label prediction for long structured texts.
Outcome: The proposed approach outperforms existing models on short-shot prompts and SFT and standard RLVR models on complex LLM prompts.
MMAC: A Multilingual, Multimodal Alignment Framework for Cultural Grounding Evaluation (2026.acl-long)

Copied to clipboard

Challenge: Existing models lack cultural alignment across modalities and languages . a new framework to assess cultural awareness across linguistics and languages is needed .
Approach: They propose a framework that integrates tri-modally aligned cultural benchmarks and a five-dimensional evaluation protocol to assess cross-country awareness disparities.
Outcome: The proposed framework assesses cultural awareness disparities across modalities and languages . it is the first dataset aligned at the input level across text, image, and speech .
Searching for Best Practices in Retrieval-Augmented Generation (2024.emnlp-main)

Copied to clipboard

Challenge: Retrieval-augmented generation (RAG) techniques have proven to be effective in integrating up-to-date information, mitigating hallucinations, and enhancing response quality, especially in specialized domains.
Approach: They propose several strategies for deploying RAG that balance performance and efficiency.
Outcome: The proposed approaches can significantly enhance question-answering capabilities and accelerate the generation of multimodal content using a “retrieval as generation” strategy.
IPR: Intelligent Prompt Routing with User-Controlled Quality-Cost Trade-offs (2025.emnlp-industry)

Copied to clipboard

Challenge: Existing systems require users to manually select models or employ rigid routing rules that fail to capture the continuous spectrum of query complexity.
Approach: They propose a quality-constrained intelligent prompt routing framework that automatically selects optimal models based on predicted response quality and user-specified tolerance levels.
Outcome: The proposed framework achieves 43.9% cost reduction while maintaining quality parity with strongest model in the Claude family and processes requests with sub-150ms latency.
V-MAGE: A Game Evaluation Framework for Assessing Vision-Centric Capabilities in Multimodal Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing static image-text benchmarks are insufficient for evaluating multimodal large language models’ dynamic perception and interactive reasoning abilities.
Approach: They propose a game-based evaluation framework to assess multimodal large language models’ visual reasoning in dynamic, continuous-space environments.
Outcome: The proposed framework systematically assesses MLLMs’ visual reasoning in dynamic, continuous-space environments.
Audio-Aware Large Language Models as Judges for Speaking Styles (2025.findings-emnlp)

Copied to clipboard

Challenge: Audio-aware large language models (ALLMs) can understand textual and non-textual information in the audio input.
Approach: They use audio-aware large language models (ALLMs) to evaluate the speaking styles of SLMs on two tasks: voice style instruction following and role-playing.
Outcome: The proposed models can understand the textual and non-textual information in the audio input and can be used as a judge to assess the speaking styles of SLMs.
GLIMPSE: Do Large Vision-Language Models Truly Think With Videos or Just Glimpse at Them? (2025.emnlp-main)

Copied to clipboard

Challenge: Existing video benchmarks often resemble image-based questions with scans of only a few key frames, without deep temporal reasoning.
Approach: They propose a video benchmark to assess whether large vision-language models can genuinely think with videos rather than perform superficial frame-level analysis.
Outcome: The proposed benchmark consists of 3,269 videos and over 4,342 highly visual-centric questions across 11 categories, including Trajectory Analysis, Temporal Reasoning, and Forensics Detection.
When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors (2026.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) perform well on table tasks, but they still make data referencing errors (DREs) prior studies have only offered limited, small-scale analyses.
Approach: They propose inference-time strategies and lightweight critics to mitigate data referencing errors.
Outcome: The proposed model achieves an average F1 score of 78.2% in detecting both in-distribution and out-of-difference DREs and assists inference for larger models.
Domain-Specific Data Generation Framework for RAG Adaptation (2026.findings-acl)

Copied to clipboard

Challenge: Retrieval-Augmented Generation (RAG) combines the language understanding and reasoning capabilities of large language models (LLMs) with external retrieval to produce domain-grounded responses.
Approach: They propose a scalable and modular data-centric framework for generating domain-grounded question–answer–context triples tailored to diverse RAG adaptation strategies.
Outcome: The proposed framework generates domain-grounded question–answer–context triples for multiple RAG adaptation strategies.
SQL-Trail: Multi-Turn Reinforcement Learning with Interleaved Feedback for Text-to-SQL (2026.acl-long)

Copied to clipboard

Challenge: Recent large language models (LLMs) have significantly improved Text-to-SQL generation, but a gap remains between AI systems and human experts on challenging benchmarks such as BIRD-Sql.
Approach: They propose a multi-turn reinforcement learning agentic framework for Text-to-SQL that uses execution feedback to iteratively refine its predictions.
Outcome: The proposed framework outperforms proprietary systems on 7B and 14B models by **5% on average, underscoring the effectiveness of interactive, agentic workflows for robust Text-to-SQL generation.
Programming over Thinking: Efficient and Robust Multi-Constraint Planning (2026.acl-long)

Copied to clipboard

Challenge: Existing large language model approaches lack flexibility in multi-constraint planning . SCOPE achieves state-of-the-art performance while lowering cost and latency .
Approach: They propose a framework that disentangles query-specific problem reasoning from generic code execution.
Outcome: The Scalable Code Planning Engine achieves state-of-the-art performance while lowering cost and latency.
NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation (2023.acl-long)

Copied to clipboard

Challenge: Existing work generates long videos segment by segment sequentially, which is inefficient.
Approach: They propose a Diffusion over Difference architecture for eXtremely Long video generation.
Outcome: The proposed architecture reduces the average inference time from 7.55min to 26s (94.26%) and generates high-quality long videos with both global and local coherence.
Learning from Textual Radiology Reports: A Benchmark Dataset for Coronary CT Angiography (2026.acl-industry)

Copied to clipboard

Challenge: CCTA reports provide an assessment of coronary disease severity to guide patient management.
Approach: They propose a pipeline that decouples structuring from classification by an LLM-based parser . CCTA-RADS is the largest publicly available dataset of CCDA reports .
Outcome: The proposed approach improves the F1-score by 6%-13% compared with direct methods.
Instructive Dialogue Summarization with Query Aggregations (2023.emnlp-main)

Copied to clipboard

Challenge: Conventional dialogue summarization methods generate summaries without considering user’s specific interests.
Approach: They propose a three-step approach to synthesize high-quality query-based summarization triples by training a unified model on three summarizing datasets with multi-purpose instructive triples.
Outcome: The proposed model outperforms state-of-the-art models and even models with larger sizes on four datasets including dialogue summarization and dialogue reading comprehension.
SeaEval for Multilingual Foundation Models: From Cross-Lingual Alignment to Cultural Reasoning (2024.naacl-long)

Copied to clipboard

Challenge: a new benchmark for multilingual foundation models is being developed . brittleness of foundation models in the dimensions of semantics and multilinguality is a key limitation .
Approach: They propose a benchmark for multilingual foundation models, SeaEval . they examine how well these models comprehend cultural practices, nuances, and values .
Outcome: The proposed model can be used to evaluate multilingual and multicultural scenarios.
Shanks: Simultaneous Hearing and Thinking for Spoken Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing large language models and spoken language models (SLMs) begin thinking and taking actions only after the user has finished their turn.
Approach: They propose a general inference framework that enables SLMs to generate unspoken chain-of-thought reasoning while listening to user input.
Outcome: The proposed framework enhances real-time user–SLM interaction in two scenarios.

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