Papers by Yanfeng Wang

37 papers
HSDreport: Heart Sound Diagnosis with Echocardiography Reports (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for heart sound diagnosis are limited to a few fixed categories and do not utilize echocardiography reports, the gold standard in the diagnosis of related diseases.
Approach: They propose a benchmark that mandates the direct utilization of heart sounds obtained from auscultation to predict echocardiography reports.
Outcome: The proposed method outperforms existing methods and existing multimodal LLMs in detecting key abnormalities in heart sounds.
RaTEScore: A Metric for Radiology Report Generation (2024.emnlp-main)

Copied to clipboard

Challenge: Existing metrics to evaluate the quality of medical reports are limited due to the complexity of clinical free-form texts.
Approach: They propose a new metric to assess the quality of medical reports generated by AI models.
Outcome: The proposed metric is based on a medical NER dataset and trained on NER models . it aligns more closely with human preference than existing metrics, the authors show .
AutoMedEval: Harnessing Language Models for Automatic Medical Capability Evaluation (2025.acl-long)

Copied to clipboard

Challenge: Existing evaluation methods based on large language models (LLMs) are expensive and lack expertise due to limitations in human expertise.
Approach: They propose an open-source automatic evaluation model with 13B parameters specifically engineered to measure the question-answering proficiency of medical LLMs.
Outcome: The proposed model surpasses baselines in terms of correlation with human judgments.
Pruning before Fine-tuning: A Retraining-free Compression Framework for Pre-trained Language Models (2024.lrec-main)

Copied to clipboard

Challenge: Structured pruning is an effective technique for compressing pre-trained language models (PLMs), but it requires retraining, leading to additional computational overhead.
Approach: They propose a task-specific pruning framework that prunes redundant modules of pre-trained language models before fine-tuning them.
Outcome: The proposed pruning framework achieves higher performance on GLUE, SQUAD, WikiText-2, Wik-103, and PTB datasets while reducing the time required for fine-tuning.
Miner: Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models (2026.acl-long)

Copied to clipboard

Challenge: Current critic-free RL methods for large reasoning models suffer from severe inefficiency when training on positive homogeneous prompts.
Approach: They propose a method that repurposes the policy’s intrinsic uncertainty as a self-supervised reward signal, with no external supervision, auxiliary models, or additional inference cost.
Outcome: Evaluated across six reasoning benchmarks on Qwen3-4B and Qwend3-8B base models, the proposed method achieves state-of-the-art performance among the other four methods.
DictLLM: Harnessing Key-Value Data Structures with Large Language Models for Enhanced Medical Diagnostics (2024.findings-acl)

Copied to clipboard

Challenge: Structured data processing is a complex and complex process.
Approach: They propose a framework that captures heterogeneity of structured data using large language models . they propose group positional encoding, hierarchical attention bias and optimal transport alignment layer .
Outcome: The proposed framework outperforms baseline methods and few-shot GPT-4 on a medical lab report dataset.
Self-Improvement of Non-autoregressive Model via Sequence-Level Distillation (2023.emnlp-main)

Copied to clipboard

Challenge: Existing non-autoregressive Transformers (NAT) models generate the entire sequence in parallel, but the multimodality problem limits their performance.
Approach: They propose a method to generate distilled data by the NAT model itself, eliminating the need for additional teacher networks.
Outcome: The proposed method can generate distilled data by the NAT model without teacher networks and adapt to different NAT models without precise adjustments.
DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought Correction (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) often prioritize reasoning over adherence to detailed instructions due to high computational costs and limited parameter access.
Approach: They propose a lightweight framework that guides small language models to refine LLMs’ outputs through chain-of-thought correction.
Outcome: The proposed framework improves the average format accuracy and content correctness of LLM outputs by 35.4% and 29.4%, respectively, achieving state-of-the-art (SOTA) performance over other competitive baselines.
SDA: Semantic Discrepancy Alignment for Text-conditioned Image Retrieval (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods for textconditioned image retrieval are limited due to the scale of training and the complexity of attributes in modification texts.
Approach: They propose a general boosting framework to address these issues by employing semantic discrepancy alignment by formulating distance consistency and neighbor consistency between the image and text domains.
Outcome: The proposed framework improves retrieval performance on three prominent datasets with state-of-the-art results.
RA2FD: Distilling Faithfulness into Efficient Dialogue Systems (2024.emnlp-main)

Copied to clipboard

Challenge: Retrieval Augmented Generation (RAG) is effective but inference inefficient, while Retrieral Free Generations (RFG) are more efficient but sacrifice faithfulness.
Approach: They propose a retrieval-free model training scheme that uses a teacher-student framework to distill the faithfulness capacity of a student's knowledge-infused responses.
Outcome: The proposed model surpasses the previous SOTA RFG model on knowledge-grounded dialogue datasets by an average of 33% while improving inference efficiency.
DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to reliability of large language models often lack self-correction or use costly post-hoc verification.
Approach: They propose a decoding framework that enhances generation reliability through real-time hallucination detection and efficient error correction.
Outcome: Extensive experiments across five benchmarks show the proposed framework improves truthfulness and factual accuracy.
CE-VDG: Counterfactual Entropy-based Bias Reduction for Video-grounded Dialogue Generation (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods to reduce question-related bias in video-grounded dialogue generation (VDG) however, the dataset often contains inherent bias, which can cause VDG models to learn spurious correlations between questions and answers.
Approach: They propose to extend the counterfactual reasoning from the information entropy perspective to the generative task, which can effectively reduce the question-related bias in the auto-regressive generation task.
Outcome: The proposed method can reduce question-related bias in the auto-regressive generation task by using counterfactual entropy as an external loss.
Bridging the Dynamic Perception Gap: Training-Free Draft Chain-of-Thought for Dynamic Multimodal Spatial Reasoning (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for dynamic spatial reasoning are limited to text or static visual domains .
Approach: They propose a framework that augments textual reasoning chains with dynamic visual drafts .
Outcome: The proposed framework outperforms existing methods in dynamic spatial reasoning tasks.
Towards Omni-RAG: Comprehensive Retrieval-Augmented Generation for Large Language Models in Medical Applications (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to source planning fail to achieve this due to misalignment between the model’s expectation of the sources and their actual content.
Approach: They propose a method to optimise large-scale medical knowledge models by combining multiple medical knowledge sources into one query.
Outcome: The proposed method significantly improves multi-source planning performance while training a smaller model to learn source alignment.
CF-TCIR: A Compositor-Free Framework for Hierarchical Text-Conditioned Image Retrieval (2024.findings-acl)

Copied to clipboard

Challenge: Existing approaches to text-conditioned image retrieval use attention-driven compositors . instead, we reformulate the retrieval process as a cross-modal interaction between a synthesized image feature and its corresponding text descriptor.
Approach: They propose a compositor-free framework for text-conditioned image retrieval . they propose combining a reference image and modification text to form a query tuple .
Outcome: The proposed framework offers advantages in terms of computational efficiency, scalability, and performance.
FGraDA: A Dataset and Benchmark for Fine-Grained Domain Adaptation in Machine Translation (2022.lrec-1)

Copied to clipboard

Challenge: Recent research on domain adaptation neglects diversity in translation within a domain . current research on NMT models considers very broad target domains .
Approach: They propose a fine-grained domain adaptation task for autonomous vehicles, AI education, real-time networks, and smart phone.
Outcome: The proposed task is compared with a dataset of Chinese-English translation tasks for four sub-domains of information technology: autonomous vehicles, AI education, real-time networks, and smart phone.
Eliciting Medical Reasoning with Knowledge-enhanced Data Synthesis: A Semi-Supervised Reinforcement Learning Approach (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods to enhance medical reasoning lack high-quality data.
Approach: They propose a medical knowledge-enhanced data Synthesis and Semi-supervised Reinforcement learning framework that uses rare disease knowledge to synthesize distribution-controllable reasoning questions.
Outcome: The proposed method outperforms existing methods across ten medical benchmarks and achieves up to 5.93% gain on rare diseases tasks.
MedCare: Advancing Medical LLMs through Decoupling Clinical Alignment and Knowledge Aggregation (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) have made significant progress in natural language understanding and generation, proving valuable especially in the medical field.
Approach: They propose a medical LLM through decoupling Clinical Alignment and Knowledge Aggregation which uses a and a to encode diverse knowledge in the first stage and filter out detrimental information.
Outcome: The proposed model achieves promising performance on over 20 medical tasks and specific medical alignment tasks.
SLoRA: Balancing Plasticity and Forgetting in Large Language Models for Continual Learning (2026.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) have achieved remarkable success across diverse tasks through large-scale pretraining.
Approach: They propose a framework that filters noisy components from LoRA updates via subspace similarity with the base model.
Outcome: The proposed framework improves accuracy by 12%, reduces forgetting by 29%, and filters out over 30% of LoRA parameters identified as noisy.
FedMABench: Benchmarking Mobile GUI Agents on Decentralized Heterogeneous User Data (2025.emnlp-main)

Copied to clipboard

Challenge: Mobile GUI agents have attracted tremendous research participation recently. traditional approaches to mobile agent training rely on centralized data collection.
Approach: They propose a benchmark for federated training and evaluation of mobile GUI agents . they find that federation algorithms consistently outperform local training .
Outcome: The first benchmark for federated training and evaluation of mobile GUI agents is released . it features 6 datasets with 30+ subsets, 8 federation algorithms, 10+ base models, and over 800 apps across 5 categories .
AgentGC: Evolutionary Learning-based Lossless Compression for Genomics Data with LLM-driven Multiple Agent (2026.findings-acl)

Copied to clipboard

Challenge: Lossless compression has made significant advancements in Genomics Data storage, sharing and management.
Approach: They propose a novel agent-based GD Compressor with 3 layers with a multi-agent named Leader and Worker.
Outcome: The proposed method improves on existing methods with low-level modeling and limited adaptability and user-unfriendly interface.
Cross-Modal Coreference Alignment: Enabling Reliable Information Transfer in Omni-LLMs (2026.acl-long)

Copied to clipboard

Challenge: Experiments on 13 Omni-LLMs reveal systematic weaknesses in cross-modal coreference . cross-module coreference is a crucial missing piece for advancing robust omni-modal reasoning.
Approach: They propose a cross-modal coreference problem to evaluate and enhance Omni-LLMs' reasoning capabilities.
Outcome: Experiments on 13 Omni-LLMs show they lack coreference-aware thinking patterns . the CROSSOMNI dataset yields significant performance gains and generalizes well to collaborative reasoning tasks.
M3AV: A Multimodal, Multigenre, and Multipurpose Audio-Visual Academic Lecture Dataset (2024.acl-long)

Copied to clipboard

Challenge: Publishing open-source academic video recordings is an emerging approach to sharing knowledge online.
Approach: They propose a multimodal, multigenre, and multipurpose audio-visual academic lecture dataset with human annotations for multimodal content recognition and understanding tasks.
Outcome: The proposed dataset can be used for multiple audio-visual recognition and understanding tasks.
ReflecTool: Towards Reflection-Aware Tool-Augmented Clinical Agents (2025.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have shown promising potential in the medical domain, assisting with tasks like clinical note generation and patient communication.
Approach: They propose a framework that excels at utilizing domain-specific tools within two stages.
Outcome: The proposed framework surpasses the pure LLMs with more than 10 points and the well-established agent-based methods with 3 points.
EvolveBench: A Comprehensive Benchmark for Assessing Temporal Awareness in LLMs on Evolving Knowledge (2025.acl-long)

Copied to clipboard

Challenge: Existing studies have explored how LLMs perceive time, but they often overlook the critical aspect of knowledge utilization.
Approach: They propose a benchmark that evaluates temporal competence along five key dimensions: Cognition, Awareness, Trustworthiness and reasoning.
Outcome: EvolveBench measures temporal competence along five key dimensions: Cognition, Awareness, Trustworthiness, Understanding and reasoning.
MatchTime: Towards Automatic Soccer Game Commentary Generation (2024.emnlp-main)

Copied to clipboard

Challenge: Existing data on soccer commentary are often unsatisfactory, and the quality of existing data is often poor.
Approach: They propose to manually annotate timestamps for 49 soccer matches and then use them to create a model to correct and filter existing data.
Outcome: The proposed model improves the viewing experience of soccer and can be trained on the curated dataset.
TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route (2025.emnlp-main)

Copied to clipboard

Challenge: Existing studies on large language models have limited evaluation of their geospatial cognition . a unified framework for evaluating geospcial cognition in LLMs remains absent .
Approach: They propose a benchmark to evaluate the geospatial route cognition of Large Language Models . they propose 'pathbuilder' tool for converting natural language instructions into navigation routes .
Outcome: The proposed framework and metrics evaluate 9 state-of-the-art LLMs on route reversal task.
Post-decoder Biasing for End-to-End Speech Recognition of Multi-turn Medical Interview (2024.lrec-main)

Copied to clipboard

Challenge: End-to-end (E2E) models are replacing hybrid models for automatic speech recognition tasks.
Approach: They propose a method to optimize E2E models for automatic speech recognition . they propose MED-IT, a multi-turn consultation speech dataset .
Outcome: The proposed method improves on subsets of rare words appearing in training speech.
KnowledgeSG: Privacy-Preserving Synthetic Text Generation with Knowledge Distillation from Server (2024.emnlp-main)

Copied to clipboard

Challenge: Existing methods to train large language models on private data are not effective because they rely on a local model for generation, resulting in a performance decline, or expose private data to API servers.
Approach: They propose a client-server framework which enhances synthetic data quality and improves model performance while ensuring privacy.
Outcome: The proposed framework improves model performance and privacy while learning local knowledge from the private data with differential privacy (DP) and distilling professional knowledge from server.
VocalNet: Speech LLMs with Multi-Token Prediction for Faster and High-Quality Generation (2025.emnlp-main)

Copied to clipboard

Challenge: Experimental results show VocalNet outperforms existing open-source speech LLMs despite limited training data.
Approach: They propose a scalable and model-agnostic training framework and a novel multi-token prediction paradigm for speech generation.
Outcome: The proposed model outperforms open-source speech LLMs while outperforming existing open-sourced models.
MM-SAP: A Comprehensive Benchmark for Assessing Self-Awareness of Multimodal Large Language Models in Perception (2024.acl-long)

Copied to clipboard

Challenge: Recent advances in multimodal large language models (MLLMs) have demonstrated exceptional capabilities in visual perception and understanding, but they also suffer from hallucinations, which limit their reliability as AI systems.
Approach: They propose a benchmark to evaluate self-awareness in perception for multimodal large language models (MLLMs) by integrating image information with knowledge quadrants, and propose MM-SAP to evaluate this capability.
Outcome: The proposed benchmark offers detailed analysis of MLLMs with self-awareness in perception.
HeteroRAG: A Heterogeneous Retrieval-Augmented Generation Framework for Medical Vision Language Tasks (2026.findings-acl)

Copied to clipboard

Challenge: Medical large vision-language models suffer from factual inaccuracies and unreliable outputs.
Approach: They propose a framework that enhances Med-LVLMs through heterogeneous knowledge sources.
Outcome: The proposed framework improves Med-LVLMs through heterogeneous knowledge sources.
MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools (2026.acl-long)

Copied to clipboard

Challenge: Existing research on Large Language Models (LLMs) relies on few servers and lacks training support.
Approach: They propose a web-agent-driven pipeline for large-scale server discovery, data synthesis, and model training that collects and filters data from 1166 servers and 11536 tools.
Outcome: Empirical evidence shows that MCP-Flow generates higher quality instruction-function call pairs and higher agentic task performance than previous work.
Synthesizing Post-Training Data for LLMs through Multi-Agent Simulation (2025.acl-long)

Copied to clipboard

Challenge: Extensive experiments demonstrate that our framework effectively generates both general and domain-specific data.
Approach: They propose a multi-agent simulator that automatically generates diverse text-based scenarios, capturing a wide range of real-world human needs.
Outcome: Experiments show that the proposed model outperforms Meta’s Llama-3-8B-Instruct model on AlpacaEval 2 and Arena-Hard benchmarks with just 20K instruction-response pairs.
When Seeing Is not Enough: Revealing the Limits of Active Reasoning in MLLMs (2026.acl-long)

Copied to clipboard

Challenge: Existing evaluations of multimodal large language models focus on passive inference, where seeing is not enough.
Approach: They propose a benchmark to evaluate active reasoning in multimodal large language models . they propose to acquire missing evidence and iteratively refine decisions under incomplete information .
Outcome: The proposed model performs better on active reasoning than on passive inference settings.
FedDQC: Data Quality Control in Federated Instruction-tuning of Large Language Models (2025.findings-acl)

Copied to clipboard

Challenge: Federated Learning (FL) enables privacy-preserving collaborative instruction tuning of large language models.
Approach: They propose a federated instruction tuning framework with dynamic data quality control to solve this problem.
Outcome: The proposed framework improves performance on mixed-quality datasets on synthetic and real-world datasets.
CliMedBench: A Large-Scale Chinese Benchmark for Evaluating Medical Large Language Models in Clinical Scenarios (2024.emnlp-main)

Copied to clipboard

Challenge: Chinese medical large language models (LLMs) are underperforming on this benchmark, especially where medical reasoning and factual consistency are vital.
Approach: They propose a benchmark with 14 expert-guided clinical scenarios to assess the medical ability of large language models across 7 pivot dimensions.
Outcome: The proposed benchmark has been validated in several ways.

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