Papers by Yihua Zhang

7 papers
SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) have highlighted the need for effective unlearning mechanisms to comply with data regulations and ethical AI practices.
Approach: They propose a second-order optimization-based LLM unlearning framework which extends the static, one-shot model update using influence unlearning to a dynamic, iterative unlearning process.
Outcome: The proposed framework outperforms first-order methods across unlearning tasks, models, and metrics.
SEUF: Is Unlearning One Expert Enough for Mixture-of-Experts LLMs? (2025.acl-long)

Copied to clipboard

Challenge: Recent advances in LLMs unlearning have shown remarkable success in removing unwanted data-model influences while preserving the model’s utility for legitimate knowledge.
Approach: They propose a Selected-Expert Unlearning Framework (SEUF) that combines expert attribution and an anchor loss to ensure controlled unlearning.
Outcome: Experiments show that the proposed framework improves forget quality and model utility by 35% on MoE LLMs across benchmarks and LLM architectures compared to standard unlearning algorithms .
Exploring Label Hierarchy in a Generative Way for Hierarchical Text Classification (2022.coling-1)

Copied to clipboard

Challenge: Existing methods for hierarchical text classification are lacking in the field of natural language processing.
Approach: They propose a hierarchy-aware T5 model with path-adaptive attention mechanism to exploit hierarchical dependency across different levels.
Outcome: The proposed model outperforms state-of-the-art models especially in Macro-F1 and low Macro.
Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods for LRM unlearning overlook critical information leakage in reasoning traces, even when final answers are successfully removed.
Approach: They propose a method that suppresses reasoning traces while preserving the model's general reasoning ability.
Outcome: The proposed method significantly reduces reasoning trace leakage and achieves strong performance across reasoning and safety benchmarks, including WMDP, StrongReject, JBB-Behaviors and WildJailbreak.
Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearning (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for unlearning in large language models often hallucinate, generate abnormal token sequences, or behave inconsistently, raising safety and trust concerns.
Approach: They propose a formal definition of unlearning honesty that preserves both utility and honesty on retained knowledge and ensures effective forgetting while encouraging the model to acknowledge its limitations.
Outcome: The proposed method achieves highest rejection rate and refusal stability on Q A tasks from the forget set, nearly double the second-best method.
ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (2026.findings-acl)

Copied to clipboard

Challenge: Recent advances in multimodal recommenders lack explicit reasoning and self-awareness of uncertainty.
Approach: They propose a reasoning-augmented multimodal agent structured around a three-stage explicit reasoning pipeline.
Outcome: The proposed agent improves ranking metrics and performance on four standard recommendation tasks across five real-world datasets.
Android in the Zoo: Chain-of-Action-Thought for GUI Agents (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing studies on large language models (LLMs) focus on the semantics of smartphone operations.
Approach: They propose a large language model (LLM) which predicts a sequence of actions of API by analyzing past actions and visual observations.
Outcome: The proposed model improves the prediction of actions on a zero-shot Android-In-The-Zoo dataset compared to previous models .

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