Papers by Yun Lu

13 papers
ACBQ: Adaptive Cross-Block Quantization of Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for post-training quantization struggle to support weight–activation joint quantization and extreme low-bit weight quantization.
Approach: They propose a framework that addresses weight–activation joint quantization and extreme weight quantization.
Outcome: The proposed framework achieves superior performance under both W4A4 and highly aggressive W2 settings while incurring negligible additional computational overhead.
mCLIP: Multilingual CLIP via Cross-lingual Transfer (2023.acl-long)

Copied to clipboard

Challenge: Existing multilingual vision-language pretrained models are biased towards English due to the lack of sufficient non-English image-text pairs.
Approach: They propose to train a retrieval-efficient dual-stream multilingual VLP model by aligning CLIP model and a multilingual text encoder through a novel Triangle Cross-modal Knowledge Distillation method.
Outcome: Empirical results show that mCLIP achieves new state-of-the-art performance for both zero-shot and finetuned multilingual image-text retrieval tasks.
Representation Potentials of Foundation Models for Multimodal Alignment: A Survey (2025.emnlp-main)

Copied to clipboard

Challenge: foundation models learn highly transferable representations through large-scale pretraining on diverse data.
Approach: They examine the representation potentials of foundation models by examining their latent capacity to capture task-specific information within a single modality while providing a transferable basis for alignment and unification across modalities.
Outcome: The foundation models exhibit remarkable similarities across architectures and modalities, the authors show . the models can capture task-specific information within a single modality while providing a transferable basis for alignment and unification across modality.
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.
Revealing the Seen, Imagining the Beyond: A Survey of Image-Grounded Chain-of-Thought Reasoning in Multimodal LLMs (2026.acl-long)

Copied to clipboard

Challenge: Recent advances in Multimodal Large Language Models (MLLMs) have shifted visual reasoning from tool-calling to end-to-end perceptionreasoning.
Approach: They synthesize the emerging paradigm of Image-Grounded Chain-of-Thought (IG-CoT) they propose a method-centric taxonomy covering prompting, supervised fine-tuning, and reinforcement learning .
Outcome: The proposed model is based on a method-centric taxonomy and benchmarks.
An Unsupervised Method for Building Sentence Simplification Corpora in Multiple Languages (2021.findings-emnlp)

Copied to clipboard

Challenge: Existing methods to build parallel sentence simplification corpora are limited . SS is used to rephrase sentences into simpler forms for those with cognitive disabilities .
Approach: They propose to build SS corpora from large-scale bilingual translation corpors using a parallel approach.
Outcome: The proposed method outperforms the existing methods on WikiLarge and achieves state-of-the-art results.
LLMs for Now, Fine-Tuning for Later: An Ensemble Approach to Data Drift in Domain-Specific Tasks (2026.acl-srw)

Copied to clipboard

Challenge: Deploying machine learning models in domain-specific scenarios is challenged by data drift and the scarcity of expert annotations.
Approach: They propose a system that combines an LLM, an AL-assisted compact model and an automatic switch module to assist the active learning process.
Outcome: The proposed system achieves 96–98% switch accuracy and outperforms both models used alone.
WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learning (2025.emnlp-main)

Copied to clipboard

Challenge: Existing work on reinforcement learning has focused on single-turn tasks such as solving math problems.
Approach: They propose a framework that learns directly from online interactions by asynchronously generating diverse trajectories, guided by binary rewards depending on task success.
Outcome: Experiments on the WebArena-Lite benchmark show that the framework outperforms state-of-the-art methods and strong proprietary models.
SeqAR: Jailbreak LLMs with Sequential Auto-Generated Characters (2025.naacl-long)

Copied to clipboard

Challenge: Existing studies have focused on the potential misuse of large language models (LLMs) however, the ability to align LLMs with human values is still vulnerable to malicious attacks.
Approach: They propose a red-teaming strategy to enhance LLM safety by using a framework to design jailbreak prompts automatically.
Outcome: The proposed framework achieves attack success rates of 88% and 60% in cold-start scenarios.
MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluation (2025.emnlp-main)

Copied to clipboard

Challenge: Existing large language model evaluation benchmarks focus on English, while current multilingual tasks lack parallel questions that specifically assess cross-lingual reasoning abilities.
Approach: They propose a comprehensive benchmark covering 29 languages, built on an English benchmark.
Outcome: The MMLU-ProX is a comprehensive benchmark covering 29 languages, built on an English benchmark.
Personalized Question Answering with User Profile Generation and Compression (2025.findings-emnlp)

Copied to clipboard

Challenge: Large language models are prone to providing “midguy” answers regardless of users’ knowledge background, thereby failing to meet each user’s personalized needs.
Approach: They propose to generate personalized answers with LLMs based on users’ past question-answering records.
Outcome: The proposed method generates personalized answers based on user's past question-answering records.
CluSanT: Differentially Private and Semantically Coherent Text Sanitization (2025.naacl-long)

Copied to clipboard

Challenge: Existing implementations of Differential Privacy (DP) in NLP typically degrade semantic integrity and readability for humans, posing significant challenges for applications requiring high-quality, coherent text processing.
Approach: They propose a text sanitization framework based on Metric Local Differential Privacy (MLDP) that uses large language models to create a set of potential substitute tokens and a parameterized cluster embedding to samaritize/substitute sensitive tokens.
Outcome: The proposed framework can be tuned with parameters such that existing state-of-the-art token sanitization algorithms can be described and improved.

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