Papers by Jianxin Wang

29 papers
Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: a new method for unlearning large language models is proposed to improve the performance of large language model models.
Approach: They propose a probability perturbation-based unlearning paradigm that allows models to forget implicit knowledge in large language models with a focus on generalisation.
Outcome: The proposed model improves unlearning vanilla target data while forgetting implicit knowledge.
MDR: Model-Specific Demonstration Retrieval at Inference Time for In-Context Learning (2024.naacl-long)

Copied to clipboard

Challenge: Existing methods for retrieval-based in-context learning ignore model biases and fail to retrieve the most appropriate demonstrations for different LLMs.
Approach: They propose a model-specific demonstration retrieval method that considers the biases of different LLMs at inference time.
Outcome: The proposed method improves performance on seen and unseen tasks with multi-scale inference LLMs by up to 41.2%.
ClusterAttn: KV Cache Compression under Intrinsic Attention Clustering (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for sparse attention apply the same pattern across different attention heads and inputs, but fail to capture the intrinsic attention clustering in large language models.
Approach: They propose a training-free sparse attention method that provides an efficient prompt cache compression scheme under intrinsic attention clustering for efficient LLM inference.
Outcome: The proposed method reduces memory usage by 10%–65% and increases throughput by 2.6–4.8 times with no accuracy loss.
Modeling Aspect Correlation for Aspect-based Sentiment Analysis via Recurrent Inverse Learning Guidance (2022.coling-1)

Copied to clipboard

Challenge: Existing methods to learn complex sentence with multiple aspects do not consider correlation between aspects to distinguish overlapped feature.
Approach: They propose a method that uses aspect correlation to improve aspect correlation modeling . they use Recurrent Mechanism to improve the joint representation of aspects .
Outcome: The proposed method is state-of-the-art in multiaspect scenarios.
Multi-Modal Knowledge Graph Transformer Framework for Multi-Modal Entity Alignment (2023.findings-emnlp)

Copied to clipboard

Challenge: Multi-modal entity alignment (MMEA) is a critical task that aims to identify equivalent entity pairs across multi-modal knowledge graphs (MMKGs).
Approach: They propose a novel MMEA transformer that hierarchically introduces neighbor features, multi-modal attributes, and entity types to enhance alignment task.
Outcome: The proposed transformer hierarchically introduces neighbor features, multi-modal attributes, and entity types to enhance the alignment task.
Unveiling Internal Reasoning Modes in LLMs: A Deep Dive into Latent Reasoning vs. Factual Shortcuts with Attribute Rate Ratio (2025.emnlp-main)

Copied to clipboard

Challenge: Existing research in multi-hop questions has identified two reasoning modes, but has not investigated how these modes differ during inference.
Approach: They propose a classification metric that compares latent reasoning and factual shortcuts in multi-hop questions.
Outcome: The proposed metric achieves 90% accuracy on the proposed datasets and demonstrates effectiveness in RAG conflict scenarios.
SSS: Editing Factual Knowledge in Language Models towards Semantic Sparse Space (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods to modify LMs suffer from sub-optimal locality, where irrelevant neighborhood examples can be adversely influenced.
Approach: They propose to use a model editing method to modify specific examples in LMs to improve locality and reasoning capability by directing the hidden state of edit example towards spaces where semantics are sparse.
Outcome: The proposed method improves locality and reasoning capability on two datasets.
DDO: Dual-Decision Optimization for LLM-Based Medical Consultation via Multi-Agent Collaboration (2025.emnlp-main)

Copied to clipboard

Challenge: Existing LLMs fail to capture the dual nature of medical consultation (MC) this mismatch often results in ineffective symptom inquiry and unreliable disease diagnosis.
Approach: They propose a novel LLM-based framework that performs Dual-Decision Optimization by decoupling the two sub-tasks and optimizing them with distinct objectives through a collaborative multi-agent workflow.
Outcome: The proposed framework outperforms existing LLM-based approaches on three real-world MC datasets and achieves competitive performance with state-of-the-art generation-based methods.
Multi-modal Concept Alignment Pre-training for Generative Medical Visual Question Answering (2024.findings-acl)

Copied to clipboard

Challenge: Medical Visual Question Answering (Med-VQA) aims to provide accurate answers to questions regarding medical images, a task particularly challenging for open-ended questions.
Approach: They propose a multi-modal concept alignment pre-training approach for generative Med-VQA that leverages a knowledge graph sourced from medical image-caption datasets and the Unified Medical Language System.
Outcome: The proposed approach significantly outperforms existing methods on a set of benchmark datasets and shows high efficiency and knowledge-image alignment capability.
medIKAL: Integrating Knowledge Graphs as Assistants of LLMs for Enhanced Clinical Diagnosis on EMRs (2025.coling-main)

Copied to clipboard

Challenge: Electronic Medical Records (EMRs) are the digitized record of a patient's medical and health information and are integral to modern healthcare.
Approach: They propose a framework that combines Large Language Models (LLMs) with knowledge graphs (KGs) to enhance diagnostic capabilities.
Outcome: The proposed framework assigns weighted importance to entities in medical records based on their type, enabling precise localization of candidate diseases within KGs.
Investigating Capsule Network and Semantic Feature on Hyperplanes for Text Classification (D19-1)

Copied to clipboard

Challenge: Various neural networks are designed for text classification on the basis of word embedding, but polysemy is a fundamental feature of the natural language, which brings challenges to text classification.
Approach: They propose to use capsule networks to construct the vectorized representation of semantics and utilize hyperplanes to decompose each capsule to acquire the specific senses.
Outcome: The proposed model extracts more discriminative semantic features and yields significant performance gain compared to baseline methods.
Dual-Gated Fusion with Prefix-Tuning for Multi-Modal Relation Extraction (2023.findings-acl)

Copied to clipboard

Challenge: Existing methods for multi-modal relation extraction lack useful visual information.
Approach: They propose a novel multi-modal relation extraction framework to capture deeper correlations of text, entity pair, and image/objects.
Outcome: The proposed framework captures the deeper correlations of text, entity pair, and image/objects, and extracts useful information.
Noise-injected Consistency Training and Entropy-constrained Pseudo Labeling for Semi-supervised Extractive Summarization (2022.coling-1)

Copied to clipboard

Challenge: Existing studies on semi-supervised learning methods focus on how to effectively utilize abundant unlabeled data.
Approach: They propose a semi-supervised consistency training method to regularize model predictions and a pseudo-labeling strategy to obtain high-confidence labels from unlabeled predictions.
Outcome: The proposed method improves extractive summarization over an insufficient labeled dataset.
ReasVQA: Advancing VideoQA with Imperfect Reasoning Process (2025.naacl-long)

Copied to clipboard

Challenge: Existing approaches to VideoQA often fail when complex reasoning or temporal relationships are involved.
Approach: They propose a method that leverages reasoning processes generated by Multimodal Large Language Models to improve VideoQA models.
Outcome: The proposed method improves VideoQA models on three benchmarks.
MARE: Multi-Aspect Rationale Extractor on Unsupervised Rationale Extraction (2024.emnlp-main)

Copied to clipboard

Challenge: Existing methods to extract text snippets from input text to support model predictions without explicit rationale annotation have limited their ability to capture meaningful internal correlations between aspects.
Approach: They propose a multi-aspect rationale extractor that extracts text snippets to support model predictions without explicit rationale annotation.
Outcome: The proposed method achieves state-of-the-art on two unsupervised rationale extraction benchmarks.
Enhancing Transformers for Generalizable First-Order Logical Entailment (2025.acl-long)

Copied to clipboard

Challenge: Moreover, transformers have demonstrated proficiency in logical reasoning over natural language.
Approach: They propose a logic-aware architecture that improves the performance in generalizable first-order logical entailment by combining distribution shifts and unseen knowledge.
Outcome: The proposed architecture outperforms methods designed specifically for knowledge graph query answering on a dataset with a large dataset.
RecStream: Graph-aware Stream Management for Concurrent Recommendation Model Online Serving (2025.coling-industry)

Copied to clipboard

Challenge: Existing systems that use recommendation models perform poorly under highly concurrent scenarios.
Approach: They propose a system that optimizes stream configurations based on model characteristics and concurrency levels.
Outcome: The proposed system outperforms existing methods under high concurrency scenarios.
VideoLLM Knows When to Speak: Enhancing Time-Sensitive Video Comprehension with Video-Text Duet Interaction Format (2025.findings-emnlp)

Copied to clipboard

Challenge: Recent studies on video large language models focus on model architectures and training datasets . interaction format between user and model is unsatisfactory for time-sensitive tasks .
Approach: They propose a video-text duet interaction format that allows for continuous playback of the video . when a text message ends, the video continues to play, similar to the alternative of two performers in a duet.
Outcome: The proposed format improves performance on time-sensitive tasks with minimal training efforts.
Towards Objective Fine-tuning: How LLMs’ Prior Knowledge Causes Potential Poor Calibration? (2025.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have enabled powerful domain-specific applications through supervised fine-tuning.
Approach: They propose a cognition-aware framework that applies targeted learning strategies according to the model’s prior knowledge to improve calibration.
Outcome: The proposed framework significantly improves calibration while maintaining performance, achieving an average 57% reduction in ECE compared to standard fine-tuning in Llama3-8B.
The Threat of PROMPTS in Large Language Models: A System and User Prompt Perspective (2025.findings-acl)

Copied to clipboard

Challenge: Prompts are essential for guiding model output and influencing content generation.
Approach: They propose to attack models with prompt leakage and prompt jailbreak attacks . they summarize the experimental setups of these methods and explore the relationship between prompt threats and prompt injection attacks.
Outcome: The proposed methods summarize the experimental setups and examine the relationship between prompt threats and prompt injection attacks.
CDA: A Contrastive Data Augmentation Method for Alzheimer’s Disease Detection (2023.findings-acl)

Copied to clipboard

Challenge: Existing methods for detecting AD are challenging and time-consuming due to lack of data and generalizability of the models.
Approach: They propose a contrastive data augmentation method which simulates the cognitive impairment of a patient by randomly deleting a proportion of text from the transcript to create negative samples.
Outcome: The proposed method achieves the best performance among language-based models on the benchmark ADReSS Challenge dataset.
Distantly Supervised Contrastive Learning for Low-Resource Scripting Language Summarization (2024.lrec-main)

Copied to clipboard

Challenge: Existing methods for code summarization are limited in resources and require atomic commands and category constraints to enhance code representations.
Approach: They propose a framework that leverages limited atomic commands and category constraints to enhance code representations.
Outcome: The proposed framework outperforms baseline methods in a number of domains and demonstrates superiority over competing frameworks.
Efficient Temporal Extrapolation of Multimodal Large Language Models with Temporal Grounding Bridge (2024.emnlp-main)

Copied to clipboard

Challenge: despite advances in multimodal large language models, the challenge of interpreting long-form videos remains a challenge . despite advancements in video-language benchmarks, the inefficiency in temporal grounding and limited pre-trained context window size remains .
Approach: They propose a framework that bootstraps MLLMs with advanced temporal grounding capabilities and broadens their contextual scope.
Outcome: The proposed framework significantly enhances the temporal capabilities of existing MLLMs.
Adversarial and Domain-Aware BERT for Cross-Domain Sentiment Analysis (2020.acl-main)

Copied to clipboard

Challenge: Cross-domain sentiment classification requires large amounts of labeled data.
Approach: They propose to apply a pre-training language model BERT on unsupervised domain adaptation . they propose to distill domain-specific features in a self-supervised way .
Outcome: The proposed model outperforms state-of-the-art methods on Amazon dataset . it can be applied to the unsupervised domain adaptation task without domain awareness .
RUIE: Retrieval-based Unified Information Extraction using Large Language Model (2025.coling-main)

Copied to clipboard

Challenge: Unified information extraction (UIE) aims to extract diverse structured information from unstructured text using a single model or framework.
Approach: They propose a framework that leverages in-context learning for efficient task generalization by combining LLM preferences with a keyword-enhanced reward model.
Outcome: The proposed framework performs better on eight held-out datasets than existing methods and instruction-tuning methods.
DisCo: Distilled Student Models Co-training for Semi-supervised Text Mining (2023.emnlp-main)

Copied to clipboard

Challenge: Existing text mining models are fine-tuned by fine-timing a large pre-trained language model (PLM) in downstream tasks.
Approach: They propose a semi-supervised learning framework for fine-tuning a cohort of small student models generated from a large pre-trained language model using knowledge distillation.
Outcome: The proposed framework outperforms baseline models on semi-supervised text classification and extractive summarization tasks while maintaining comparable performance.
Example Quality Matters: Multi-Aspects Example Augmentation for Private Library Programming (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to code generation fail to consider the quality of retrieved examples.
Approach: They propose a retrieval-augmented generation method that combines existing API examples to improve complexity and readability.
Outcome: The proposed method achieves up to 22% accuracy improvement over baseline methods.
Capsule Network with Interactive Attention for Aspect-Level Sentiment Classification (D19-1)

Copied to clipboard

Challenge: Existing methods for aspect-level sentiment classification are limited for dealing with overlapped features.
Approach: They propose to use capsule network to construct vector-based feature representation and cluster features by an EM routing algorithm to model semantic relationship between aspect terms and context.
Outcome: The proposed model achieves state-of-the-art on three datasets.
ModularMoE: Fast LLM Customization with Parameter-Sharing Mixture-of-Experts for Low-Resource Settings (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models impose significant computational and storage burdens on personal devices . existing customization approaches incur excessive computational costs or lead to suboptimal performance .
Approach: They propose a training framework that converts pre-trained LLMs into parameter-sharing MoE models for lightweight deployment.
Outcome: The proposed training framework outperforms state-of-the-art training frameworks at the same sparsity level while delivering up to 2.71 inference speedup.

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