Papers by Junjie Jiang

22 papers
WhiteningBERT: An Easy Unsupervised Sentence Embedding Approach (2021.findings-emnlp)

Copied to clipboard

Challenge: Pre-trained language models perform well on learning sentence semantics when fine-tuned with supervised data.
Approach: They conduct a thorough examination of pretrained model based unsupervised sentence embeddings.
Outcome: The proposed approach improves on whitening-based vector normalization with less than 10 lines of code.
Muse: Towards Reproducible Long-Form Song Generation with Fine-Grained Style Control (2026.findings-acl)

Copied to clipboard

Challenge: Recent commercial systems such as Suno demonstrate strong capabilities in long-form song generation, but academic research remains non-reproducible due to the lack of publicly available training data.
Approach: They propose a system for long-form song generation with fine-grained style conditioning that includes a licensed synthetic dataset and a song generation model, Muse.
Outcome: The proposed system achieves competitive performance on phoneme error rate, text–music style similarity, and audio aesthetic quality while enabling controllable segment-level generation across different musical structures.
DentalGPT: Incentivizing Multimodal Reasoning in Dentistry (2026.findings-acl)

Copied to clipboard

Challenge: Current multimodal large language models (MLLMs) show limited understanding of dental images.
Approach: They propose a dental-specialized multimodal large language model trained via staged multimodal alignment and reinforcement learning.
Outcome: The proposed model outperforms state-of-the-art models on disease classification and dental VQA tasks.
CoSQA: 20,000+ Web Queries for Code Search and Question Answering (2021.acl-long)

Copied to clipboard

Challenge: Using deep neural networks to find codes is difficult . we present a dataset that includes 20,604 labels for natural language queries and codes .
Approach: They introduce a contrastive learning method to enhance text-code matching . they find that CoSQA improves the accuracy of code question answering by 5.1% .
Outcome: The proposed method improves the accuracy of code question answering by 5.1% and improves by 10.5% on a CodeBERT model.
Generative Text-to-Image Retrieval via Hierarchical Identifiers and Semantic Internalization (2026.findings-acl)

Copied to clipboard

Challenge: Existing text-to-image retrieval methods suffer from limited semantic discriminability, alignment bias, and closed-set restrictions.
Approach: They propose a framework for semantic internalization for Generative Multimodal Alignment . they construct multi-granularity hierarchical identifiers to ensure unique, semantically consistent image representations .
Outcome: The proposed framework outperforms state-of-the-art frameworks on Flickr30K and MS-COCO datasets . it achieves average Recall@1, Recall @5, and Recall_10 improvements of 10.65%, 8.50%, and 7.00% .
ArchiDocGen: Multi-Agent Framework for Expository Document Generation in the Architectural Industry (2025.acl-industry)

Copied to clipboard

Challenge: drafting method statements is labor-intensive and time-consuming . traditional methods involve using static templates filled in manually by engineers .
Approach: They propose a framework that automates method statement generation by using multi-agent collaboration.
Outcome: The proposed framework achieves 4.38 ContentScore, excelling in specialization, completeness, organization, and clarity.
Benchmarking Machine Translation with Cultural Awareness (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing studies on terminology translation focus on breaking language barriers rather than cultural barriers.
Approach: They propose a parallel corpus enriched with CSI annotations in 6 language pairs for investigating Cultural-Aware Machine Translation.
Outcome: The proposed corpus is enriched with CSI annotations in 6 languages and measures translation quality.
PHMOSpell: Phonological and Morphological Knowledge Guided Chinese Spelling Check (2021.acl-long)

Copied to clipboard

Challenge: False gram and phonological errors make Chinese spelling check difficult . a novel end-to-end trainable model outperforms existing methods .
Approach: They propose a trainable Chinese spelling check model that integrates phonological and visual information into a pre-trained language model.
Outcome: The proposed model outperforms existing state-of-the-art models on three benchmarks.
Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to attack Large Language Model (LLM) tool-learning systems are black-box oriented and rely on static commands that cannot adapt flexibly to the changes in user queries and the invocation chain.
Approach: They propose a dynamic attack comment generation approach for information theft attacks in LLM tool-learning systems that mimics the familiar by inferring the information utilized by upstream tools.
Outcome: The proposed approach outperforms baselines with +13.2% ASRTheft and can be generalized to new tool-learning systems to expose their information leakage risks.
Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent Collaboration (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing studies have found that when LLMs are given criminal facts and legal rules, then asked whether cases constitute a certain charge, they struggle to understand legal theories and perform basic legal reasoning tasks.
Approach: They propose a task to assess LLMs' understanding of legal theories and reasoning capabilities by using a novel framework: Multi-Agent framework for improving complex legal reasoning capability.
Outcome: The proposed framework improves LLMs' understanding of legal theories and reasoning abilities in real-world scenarios.
From Heads to Neurons: Causal Attribution and Steering in Multi-Task Vision–Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing models focus on single tasks, limiting comparability of neuron importance . ranking strategies overlook how task-dependent information pathways shape write-in effects of feed-forward network (FFN) neurons.
Approach: They propose a gradient-free framework for task-aware neuron attribution and steering in multi-task vision-language models.
Outcome: The proposed framework outperforms existing methods in identifying task-critical neurons and improves model performance after steering.
LeCoDe: A Benchmark Dataset for Interactive Legal Consultation Dialogue Evaluation (2026.acl-long)

Copied to clipboard

Challenge: Current systems for legal consultation are insufficient to handle the knowledge-intensive nature of real-world consultations.
Approach: They propose a multi-turn benchmark dataset to evaluate LLMs in legal consultation settings.
Outcome: The proposed framework assesses LLMs’ consultation capabilities in terms of (1) clarification capability and (2) professional advice quality.
One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing knowledge poisoning attacks against RAG systems require multiple poisoned documents or can only function effectively on simplistic queries.
Approach: They propose a more realistic knowledge poisoning attack that poisons only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements.
Outcome: The proposed attack achieves success by poisoning only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements.
V-SEAM: Visual Semantic Editing and Attention Modulating for Causal Interpretability of Vision-Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing work on causal interpretability focuses on large language models (LLMs) but internal mechanisms of vision-language models remain underexplored, authors say .
Approach: They introduce a framework that combines visual and semantic manipulations for causal interpretation of vision-language models.
Outcome: The proposed framework shows improved performance for LLAVA and InstructBLIP on three diverse benchmarks.
All Changes May Have Invariant Principles: Improving Ever-Shifting Harmful Meme Detection via Design Concept Reproduction (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for harmful meme detection only learn the combination of harmful elements and lack understanding of these implicit expressions.
Approach: They propose a method that detects harmful memes by replicating the design concept of malicious users.
Outcome: The proposed method achieves the highest accuracy with 81.1% and has slight accuracy decreases when generalized to type-shifting and temporal-evolving memes.
Know Thy Enemy: Securing LLMs Against Prompt Injection via Diverse Data Synthesis and Instruction-Level Chain-of-Thought Learning (2026.findings-acl)

Copied to clipboard

Challenge: Large language model (LLM)-integrated applications face security vulnerabilities from prompt injection (PI) attacks.
Approach: They propose a model enhancement method that synthesizes diverse training data and employs instruction-level chain-of-thought fine-tuning to enable LLMs to effectively identify and reject malicious instructions regardless of their source or position in the context.
Outcome: The proposed method outperforms baselines in three critical dimensions while maintaining utility performance without degradation.
SAGE: Synergistic Adaptive Gating of Experts for Hateful Video Detection (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for hateful video detection rely on multimodal feature fusion . existing methods rely only on blind feature mixing, which leads to feature dilution .
Approach: They propose a framework that shifts from blind feature mixing to decision-level arbitration . it instantiates disentangled experts to rigorously preserve modality-specific semantics .
Outcome: The proposed framework outperforms state-of-the-art methods on HateMM and MultiHateClip benchmarks.
Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build Environments (2026.findings-acl)

Copied to clipboard

Challenge: Currently, there are no efficient reinforcement learning (RL) frameworks specifically designed for tool use.
Approach: They propose an automated environment construction pipeline that incorporates scenario decomposition, document generation, function integration, complexity scaling, and localized deployment to enable high-quality training environments without external tools.
Outcome: The proposed framework significantly improves the models’ tool-use performance without degrading their general capabilities.
REO-Relevance, Extraness, Omission: A Fine-grained Evaluation for Image Captioning (D19-1)

Copied to clipboard

Challenge: Existing metrics for image captioning evaluation provide an overall quality score, which is difficult to infer specific description errors.
Approach: They propose a fine-grained evaluation method REO for automatically measuring the performance of image captioning systems.
Outcome: The proposed method achieves higher consistency with human judgments and provides more intuitive evaluation results than other metrics.
SimpleDeepSearcher: Deep Information Seeking via Web-Powered Reasoning Trajectory Synthesis (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to deep search training lack high-quality training trajectories, prohibitive computational costs and lack of high-fidelity training data.
Approach: They propose a framework that synthesizes high-quality training data by simulating real user interactions in live web search environments.
Outcome: The proposed framework synthesizes high-quality training data by simulating user interactions in live web search environments.
Mixed-modality Representation Learning and Pre-training for Joint Table-and-Text Retrieval in OpenQA (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing methods to retrieve evidences from corpus are difficult due to table-text discrepancy and data sparsity problem.
Approach: They propose an optimized OpenQA Table-Text Retriever to retrieve tabular and textual evidences from tabular resources.
Outcome: The proposed OpenQA Table-Text Retriever significantly outperforms existing methods on QA tasks.
Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training (2026.acl-long)

Copied to clipboard

Challenge: Existing methods to predict performance of large language models are lacking . authors propose a size-dependent mutual information predictor for closed-book question answering accuracy .
Approach: They propose a size-dependent mutual information predictor that integrates knowledge frequency, knowledge specificity, and model size to forecast closed-book question answering accuracy.
Outcome: The proposed method outperforms baseline models and achieves R2 > 0.7 in predicting QA accuracy without additional training.

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