Papers by Haiyan Wu

9 papers
The Elephant in the Room: Exploring the Role of Neutral Words in Language Model Group-Agnostic Debiasing (2025.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) are increasingly integrated into our daily lives, raising ethical concerns, especially about perpetuating stereotypes.
Approach: They propose a method that incorporates a neutral word semantics-based loss function to alleviate the deterioration of the LMS during debiasing.
Outcome: The proposed method alleviates the deterioration of the Language Modeling Score (LMS) by incorporating a neutral word semantics-based loss function.
Towards Boosting LLMs-driven Relevance Modeling with Progressive Retrieved Behavior-augmented Prompting (2025.coling-industry)

Copied to clipboard

Challenge: Existing approaches to relevance modeling have lacked generalization and accuracy . recent studies have focused on capturing the semantic relationships between queries and items .
Approach: They propose a framework that integrates world knowledge stored in LLMs with specialized domain knowledge represented by user behavior data for promising performance.
Outcome: The proposed framework can handle full-scale search traffics of Alipay with acceptable cost and latency.
LLMs Trust Humans More, That’s a Problem! Unveiling and Mitigating the Authority Bias in Retrieval-Augmented Generation (2025.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) generate outputs that stray from user input or contravene established knowledge.
Approach: They propose a new phenomenon, Authority Bias, where LLMs favor one knowledge source over the other . they propose atomic information that generates conflicts and a Conflict Detection Enhanced Query framework .
Outcome: The proposed framework reduces Authority bias in large language models . it detects conflicts, performs credibility assessment on conflicting paragraphs, and detects perturbed text .
A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models (2025.findings-emnlp)

Copied to clipboard

Challenge: Sparse Autoencoders (SAEs) can disentangle complex features into more interpretable components.
Approach: They propose to use Sparse Autoencoders to disentangle LLM features into more interpretable components.
Outcome: The proposed method disentangles complex features into more interpretable components.
Can LLMs Hear the Dogwhistle? (2026.findings-acl)

Copied to clipboard

Challenge: Existing safety benchmarks focus on explicitly harmful content, but ignore context-dependent expressions such as dogwhistles.
Approach: They propose a benchmark for evaluating LLM safety under dogwhistle-driven prompts . their findings expose a blind spot in current safety evaluation practices .
Outcome: The proposed benchmark compared safety performance with toxic terms using dogwhistle-driven prompts.
Reflective RAG: Self-Evaluation Driven Strategy Optimization in Agentic Retrieval-Augmented Generation (2026.findings-acl)

Copied to clipboard

Challenge: Recent agentic RAG systems lack the capacity to evaluate the utility of retrieved information, leading to brittle reasoning and suboptimal decision-making.
Approach: They propose a framework that integrates self-evaluation to dynamically optimize retrieval and generation strategy.
Outcome: The proposed framework outperforms strong agentic baselines on five knowledge-intensive QA benchmarks and improves training stability and generalization to multi-hop reasoning tasks.
Beyond Input Activations: Identifying Influential Latents by Gradient Sparse Autoencoders (2025.emnlp-main)

Copied to clipboard

Challenge: Sparse Autoencoders (SAEs) have recently emerged as powerful tools for interpreting and steering the internal representations of large language models (LLMs).
Approach: They propose a method that identifies the most influential latents by incorporating output-side gradient information.
Outcome: The proposed method identifies the most influential latents by incorporating output-side gradient information.
Denoising Concept Vectors with Sparse Autoencoders for Improved Language Model Steering (2026.findings-eacl)

Copied to clipboard

Challenge: Existing methods for steering concept vectors suffer from noisy features in diverse datasets that undermine steering robustness.
Approach: They propose a Sparse Autoencoder-Denoised Concept Vector (SDCV) which selectively keeps the most discriminative SAE latents while reconstructing hidden representations.
Outcome: The proposed method improves steering success rates by 4-16% across six challenging concepts while maintaining topic relevance.
Modularized Syntactic Neural Networks for Sentence Classification (2020.emnlp-main)

Copied to clipboard

Challenge: Existing models for sentence classification use local information of sub-trees, but new models use global context .
Approach: They propose a tree-parallel mini-batch strategy for efficient training and predicting sentences . they propose to use syntax category labels to model sub-trees .
Outcome: The proposed model outperforms state-of-the-art tree-based methods on the sentence classification task.

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