Papers by Chuan Wu

18 papers
MACS: Modality-Aware Capacity Scaling for Efficient Multimodal MoE Inference (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for expert parallelism inference suffer from a significant efficiency bottleneck . existing methods fail to address information heterogeneity and modality dynamics .
Approach: They propose a training-free inference framework that scales experts without training . they propose an Entropy-Weighted Load mechanism to quantify the semantic value of visual tokens .
Outcome: Experiments show that MACS outperforms existing methods on multimodal benchmarks.
How Well Do LLMs Handle Cantonese? Benchmarking Cantonese Capabilities of Large Language Models (2025.findings-naacl)

Copied to clipboard

Challenge: Cantonese has scant representation in NLP research, especially compared to other languages from similarly developed regions.
Approach: They propose to evaluate Cantonese LLM performance in factual generation, mathematical logic, complex reasoning, and general knowledge in Cantonesian.
Outcome: The proposed models will evaluate Cantonese's performance in factual generation, mathematical logic, complex reasoning, and general knowledge in Cantone.
LoopCoder: Scaling Code Intelligence via Looped Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Large language models have mastered syntax-level code generation, but complex algorithmic reasoning remains a challenge.
Approach: They propose a recurrent inductive bias that aligns with the recursive nature of programming logic.
Outcome: The proposed model achieves comparable performance to standard dense models with more parameters.
WN-Salience: A Corpus of News Articles with Entity Salience Annotations (2020.lrec-1)

Copied to clipboard

Challenge: Existing work on entity salience does not distinguish between salient and non-salient entities.
Approach: They propose a dataset to measure entity salience using WikiNews dataset . WN-Salience is built on top of Wikinews, a Wikimedia project .
Outcome: The proposed dataset can be used to benchmark tasks such as entity salience detection and salient entity linking.
UniConv: Unifying Retrieval and Response Generation for Large Language Models in Conversations (2025.acl-long)

Copied to clipboard

Challenge: Existing conversational search systems are usually built with two different models . this separation restricts the system from leveraging the model's intrinsic knowledge simultaneously . Existing studies for developing unified models cannot fully address the aspects of understanding conversational context, managing retrieval independently, and generating responses.
Approach: They propose to unify dense retrieval and response generation for large language models in conversation by fine-tuning and mitigating data discrepancy.
Outcome: The proposed model can outperform existing models on five conversational search datasets and reduce inconsistency risks while mitigating data discrepancy.
Developing and Utilizing a Large-Scale Cantonese Dataset for Multi-Tasking in Large Language Models (2025.findings-emnlp)

Copied to clipboard

Challenge: Cantonese is considered a low-resource language due to the dominance of Mandarin . rich colloquial vocabulary of Cantone, English loanwords, and code-switching characteristics add to the complexity of corpus collection and processing.
Approach: We collect Cantonese texts from open source corpora, Hong Kong-specific forums, Wikipedia . we refine the model through supervised fine-tuning on curated Cantonesian tasks .
Outcome: The model achieves state-of-the-art (SOTA) performance on four Cantonese benchmarks.
Legal Fact Prediction: The Missing Piece in Legal Judgment Prediction (2025.emnlp-main)

Copied to clipboard

Challenge: Existing studies use legal facts to predict judgments, but legal facts are difficult to obtain in early stages of litigation.
Approach: They propose a legal fact prediction task that takes evidence from trial as input to make predictions in the absence of ground-truth legal facts.
Outcome: The proposed task can predict court rulings without ground-truth legal facts . the first benchmark dataset, LFPBench, is used to evaluate the task .
Exploring the Potential of Large Language Models for Heterophilic Graphs (2025.naacl-long)

Copied to clipboard

Challenge: Existing approaches for heterophilic graphs overlook rich textual data associated with nodes, which could unlock deeper insights into their heterophilistic contexts.
Approach: They propose a two-stage framework to enhance node classification on heterophilic graphs by leveraging open-world knowledge encoded by large language models.
Outcome: The proposed framework can be used to better characterize heterophilic graphs, where neighboring nodes often exhibit different labels.
A Cognitive Stimulation Dialogue System with Multi-source Knowledge Fusion for Elders with Cognitive Impairment (2023.acl-long)

Copied to clipboard

Challenge: Existing cognitive stimulation systems lack data on how to integrate emotional support and therapy principles into chit-chat dialogue systems.
Approach: They propose a multi-source knowledge fusion method for CS dialogue to generate open-ended responses guided by the therapy principle and emotional support strategy.
Outcome: The proposed method generates open-ended responses guided by the therapy principle and emotional support strategy of the target response.
LoRA Meets Dropout under a Unified Framework (2024.findings-acl)

Copied to clipboard

Challenge: Parameter-efficientfinetuning (PEFT) has gained popularity as a lightweight approach for model customization.
Approach: They propose a parameter-efficient dropout method that is overfitting-prone and parameter-freezed.
Outcome: The proposed method is superior to existing methods and compares with transformer-specific methods.
QSpec: Speculative Decoding with Complementary Quantization Schemes (2025.emnlp-main)

Copied to clipboard

Challenge: Quantization is widely adopted to accelerate inference and reduce memory consumption in large language models.
Approach: They propose a quantization paradigm that decouples efficiency from quality by integrating two complementary schemes via speculative decoding.
Outcome: The proposed approach achieves 1.64x speedup without quality degradation and outperforms state-of-the-art speculative decoding methods by 1.55x in batched settings.
Learning What Matters: Dynamic Dimension Selection and Aggregation for Interpretable Vision-Language Reward Modeling (2026.acl-long)

Copied to clipboard

Challenge: Existing multimodal reward models are interpretable but slow, while discriminative ones are opaque "black boxes."
Approach: They propose a framework that dynamically decomposes evaluation into granular, interpretable dimensions.
Outcome: The proposed framework outperforms open-source reward models on benchmarks like VL-RewardBench.
ProReason: Multi-Modal Proactive Reasoning with Decoupled Eyesight and Wisdom (2025.emnlp-main)

Copied to clipboard

Challenge: Large vision-language models often prioritize language knowledge over image information on visual reasoning tasks, incurring performance degradation.
Approach: They propose a visual reasoning framework that decouples vision-reasoning capabilities and multi-run proactive perception.
Outcome: The proposed framework outperforms existing models on benchmarks for open-source and closed-source models with 13.2% performance gain.
Unveiling Multimodal Processing: Exploring Activation Patterns in Multimodal LLMs for Interpretability and Efficiency (2025.findings-emnlp)

Copied to clipboard

Challenge: Recent advances in multimodal large language models have remained opaque.
Approach: They propose a method to convert dense MLLMs into fine-grained Mixture-of-Experts architectures.
Outcome: The proposed method outperforms random expert pruning and sparse activation and model pruning.
CIA: Inferring the Communication Topology from LLM-based Multi-Agent Systems (2026.acl-long)

Copied to clipboard

Challenge: LLM-based multi-agent systems (MAS) have demonstrated remarkable capabilities in solving complex tasks.
Approach: They propose a communication inference attack that constructs new adversarial queries to induce intermediate agents’ reasoning outputs and models their semantic correlations through the global bias disentanglement and LLM-guided weak supervision.
Outcome: The proposed attack achieves an average AUC of 0.87 and a peak AUC up to 0.99, revealing the privacy risk in MAS.
Icon2: Aligning Large Language Models Using Self-Synthetic Preference Data via Inherent Regulation (2025.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) require high quality preference datasets to align with human preferences.
Approach: They propose a framework that leverages inherent regulation of LLMs’ representation space for efficient and tailored preference dataset construction, named Icon2.
Outcome: The proposed framework improves performance on benchmarks like AlpacaEval 2.0 and Arena-Hard while reducing computational costs by up to 48.1%.
Shall We Team Up: Exploring Spontaneous Cooperation of Competing LLM Agents (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) are increasingly used in social simulations, where they are guided by carefully crafted instructions to exhibit human-like behaviors.
Approach: They propose to use Large Language Models (LLMs) as agents to simulate the gradual transition from non-cooperative to cooperative behaviors of agents.
Outcome: The proposed model can simulate the gradual transition from non-cooperative to cooperative behaviors in three competitive scenarios.
PRoLoRA: Partial Rotation Empowers More Parameter-Efficient LoRA (2024.acl-long)

Copied to clipboard

Challenge: Partially Rotation-enhanced Low-Rank Adaptation (PRoLoRA) is an intra-layer sharing mechanism that circumvents the drawbacks of peer parameter-sharing methods.
Approach: They propose a partially rotation-enhanced low-rank adaptation (PRoLoRA) that shares four components to reduce the cost of LoRA and improves model capacity.
Outcome: Empirical results show that PRoLoRA outperforms LoRA on multiple instruction tuning datasets.

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