Papers by Qinghua Li

9 papers
DiSec: Mitigating Backdoors in Pre-trained Language Models via Disentanglement of Adversarial Weights for Secure Fine-Tuning (2026.findings-acl)

Copied to clipboard

Challenge: Existing defenses rely on privileged assumptions, limiting their applicability in realistic settings.
Approach: They propose a task-agnostic backdoor attack that contaminates pre-trained language models . authors propose auxiliary text purification framework that uses only clean auxiliary data .
Outcome: The proposed framework suppresses attack success while preserving clean-task utility.
Improving Sequential Model Editing with Fact Retrieval (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing methods to fix erroneous knowledge in Pre-trained Language models experience a performance decline when the number of edits increases.
Approach: They propose a framework that leverages factual information to enhance editing generalization and guide the identification of edits by retrieving related facts from the fact-patch memory.
Outcome: The proposed framework can improve model generalization and accuracy even with thousands of edits.
Evaluating the Long-Term Memory of Large Language Models (2025.findings-acl)

Copied to clipboard

Challenge: Recent studies have not thoroughly investigated the memory performance of large language models in long-term tasks.
Approach: They propose a dataset to evaluate the long-term memory capabilities of large language models.
Outcome: The proposed model exhibits memory preferences across different categories of information.
BIC: Twitter Bot Detection with Text-Graph Interaction and Semantic Consistency (2023.acl-long)

Copied to clipboard

Challenge: Existing methods to identify bots rely on text or networks alone . text-graph interactions and semantic consistency are essential improvements to combat bot evolution.
Approach: They propose to combine text-graph interaction and semantic Consistency to model Twitter bots' behavior based on attention weights and a text-graphic interaction module to enable information exchange across modalities in the learning process.
Outcome: The proposed framework outperforms state-of-the-art methods on two widely adopted datasets and the results are consistent with previous work.
A Layer-wise Analysis of Supervised Fine-Tuning (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for fine-tuning ignore depth-dependent heterogeneity of instruction-following . a critical gap remains in understanding where these changes occur across the model's depth and which layers are essential for instruction- following.
Approach: They propose a method which selectively updates critical intermediate layers . they show that effective alignment is architecturally localized rather than distributed .
Outcome: The proposed method outperforms standard LoRA up to 10.2% on GSM8K with reduced parameter overhead.
Inference Helps PLMs’ Conceptual Understanding: Improving the Abstract Inference Ability with Hierarchical Conceptual Entailment Graphs (2024.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to abstract inference ignore the *polysemy* and *hierarchical nature of concepts* . prevailing approaches disregard how arguments might entail differently across various concept levels, thereby missing potential enlargement connections.
Approach: They propose a framework that organizes arguments hierarchically and delves into entailment relations at diverse concept levels.
Outcome: The proposed framework improves the model's generalization and reasoning prowess in natural language inference.
AlphaQT-Bench: Diagnosing the Gap between Financial Code Generation and Quantitative Reasoning in LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Existing benchmarks rely on outcome-driven metrics such as profitability and look-ahead bias.
Approach: They propose a diagnostic benchmark for instruction-grounded financial code generation under strict semantic and temporal constraints.
Outcome: The proposed benchmarks show that the models fail under causal, structural, or functional constraints.
A Knowledge-Guided Framework for Frame Identification (2021.acl-long)

Copied to clipboard

Challenge: Existing frameworks for frame identification are limited to only a few types of frame knowledge.
Approach: They propose a Knowledge-Guided Frame Identification framework that integrates frame knowledge to learn better frame representation.
Outcome: The proposed framework outperforms the state-of-the-art methods on two benchmark datasets.
How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation (2026.findings-acl)

Copied to clipboard

Challenge: Chain-of-Thought (CoT) prompting significantly enhances model reasoning, yet its internal mechanisms remain poorly understood.
Approach: They reversely traced information flow across decoding, projection, and activation phases and found that CoT may serve as a decoding space pruner .
Outcome: The proposed framework can be used to design more efficient and robust prompts.

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