Papers by Pan Xie

21 papers
Exploring Question-Specific Rewards for Generating Deep Questions (2020.coling-main)

Copied to clipboard

Challenge: Recent question generation approaches use the sequence-to-sequence framework to optimize the log likelihood of ground-truth questions using teacher forcing.
Approach: They propose to optimize for QG-specific objectives via reinforcement learning to improve question quality.
Outcome: The proposed model improves the fluency, relevance, and answerability of generated questions.
MLLM-Protector: Ensuring MLLM’s Safety without Hurting Performance (2024.emnlp-main)

Copied to clipboard

Challenge: MLLMs are deployed on limited image-text pairs, which makes them more vulnerable to catastrophic forgetting of their original abilities during safety fine-tuning.
Approach: They propose a plug-and-play strategy that detects harmful visual inputs and transforms harmful ones into harmless ones.
Outcome: The proposed approach mitigates the risks posed by malicious visual inputs without compromising the original performance of MLLMs.
Inductive Topic Variational Graph Auto-Encoder for Text Classification (2021.naacl-main)

Copied to clipboard

Challenge: Existing methods for text classification do not assume explicit latent semantic structure of documents, making them less effective and difficult to interpret.
Approach: They propose a model that integrates a topic model into variational graph-auto-encoder to capture hidden semantic information between documents and words.
Outcome: The proposed model outperforms existing models on supervised and semi-supervised text classification and unsupervised representation learning.
Graph Relational Topic Model with Higher-order Graph Attention Auto-encoders (2021.findings-acl)

Copied to clipboard

Challenge: Existing methods for learning low-dimensional representations of networked documents are largely ignored for document networks.
Approach: They propose a graph relational topic model to explore document neighborhood information . the model can learn efficient networked document representations in the latent topic space .
Outcome: The proposed model outperforms existing methods on unsupervised representation learning and other downstream tasks.
MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues (2026.findings-acl)

Copied to clipboard

Challenge: Existing evaluation benchmarks for Multimodal Large Language Models (MLLMs) focus on single-turn question answering, overlooking the complexity of multi-turn dialogues in real-world scenarios.
Approach: They propose a video understanding benchmark for MLLMs in multi-turn dialogues that assesses six core competencies that focus on perceptivity and interactivity.
Outcome: The MT-Video-Bench evaluates 1,000 multi-turn dialogues from diverse domains and reveals significant performance discrepancies and limitations in handling multi-turned video dialogues.
PlanningArena: A Modular Benchmark for Multidimensional Evaluation of Planning and Tool Learning (2025.acl-long)

Copied to clipboard

Challenge: Recent studies have shown that LLMs can be significantly improved by integrating external tools.
Approach: They propose a framework that integrates external tools into large language models to evaluate their ability to generate action plans.
Outcome: The proposed framework evaluates the ability of large language models to generate action plans and generate action plan templates.
Making LLMs Better Many-to-Many Speech-to-Text Translators with Curriculum Learning (2025.acl-long)

Copied to clipboard

Challenge: Existing studies on English-centric translation tasks have focused on multimodal large language models, but the exploration of many-to-many translation is limited by the scarcity of parallel data.
Approach: They propose a three-stage curriculum learning strategy that leverages the machine translation capabilities of large language models and adapts them to S2TT tasks.
Outcome: The proposed strategy achieves state-of-the-art average performance in 1514 language pairs, requiring fewer than 10 hours of speech data per language to achieve competitive results.
AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge (2025.acl-long)

Copied to clipboard

Challenge: Existing studies solve this challenge by updating benchmarks with newly collected data, but they fail to guarantee contamination-free evaluation as the newly collected knowledge may contain pre-existing knowledge.
Approach: They propose an automated anti-leakage benchmarking framework that builds and updates benchmarks without human labor instead of using newly collected data.
Outcome: The proposed framework significantly reduces the cost of benchmark maintenance to accommodate emerging LLMs.
User-Assistant Bias in LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Modern large language models are typically trained using structured role tags . asymmetries in training data associated with different role tags can potentially introduce inductive biases.
Approach: They propose a task-agnostic benchmark to test user–assistant bias in large language models . they find human-preference alignment amplifies user bias, reasoning fine-tuning reduces it .
Outcome: The proposed benchmark tests show that most instruction-tuned models exhibit strong user bias . human-preference alignment amplifies user bias, while reasoning fine-tuning reduces it.
MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus (2026.findings-acl)

Copied to clipboard

Challenge: Multimodal Large Language Models (MLLMs) have advanced Chinese Classical Studies (CCS) but the audio dimension of CCS remains underexplored due to a lack of high-quality, domain-specific audio corpora.
Approach: They propose a 119-hour audio corpus comprising 22,000 audio samples to bridge this gap . it encompasses a diverse range of literary genres across six tasks .
Outcome: The proposed corpus encompasses a diverse range of literary genres across six tasks: Automatic Speech Recognition (ASR), Speech-to-Text Translation (S2TT), Speech Emotion Captioning (SEC), Spoken Question Answering ( SQA), Speech Understanding (SU), and Speech Reasoning (SR).
GOBench: Stage-Wise Diagnostics and the Visual Paradox in Multimodal Graph Optimization (2026.findings-acl)

Copied to clipboard

Challenge: Existing benchmarks fail to represent multimodal problem specifications, score outcomes only and cannot localize where failures occur along the modeling pipeline.
Approach: They propose a Graph Optimization benchmark that aligns multiple modalities with solver-derived oracles and a diagnostic protocol that evaluates intermediate artifacts as well as end results.
Outcome: Graph Optimization benchmark (GOBench) evaluates intermediate artifacts as well as end results . vision reliably increases inference cost, while reliability impact is regime-dependent . current benchmarks fail to represent multimodal problem specifications, fail to localize failures .
DaMo: Data Mixing Optimizer in Fine-tuning Multimodal LLMs for Mobile Phone Agents (2026.findings-acl)

Copied to clipboard

Challenge: Mobile Phone Agents (MPAs) have attracted huge attention due to their practicability in a multitude of scenarios.
Approach: They propose a data mixture optimization solution that extrapolates optimal data mixtures from a trainable network.
Outcome: The proposed model outperforms existing methods on open-source benchmarks and on open source benchmarks.
IGenBench: Benchmarking the Reliability of Text-to-Infographic Generation (2026.acl-long)

Copied to clipboard

Challenge: Generated infographics may appear correct at first glance but contain easily overlooked issues, such as distorted data encoding or incorrect textual content.
Approach: They propose to evaluate reliability of text-to-infographic generation using IGenBench . they employ multimodal large language models to verify each question .
Outcome: The proposed framework decomposes reliability verification into atomic yes/no questions based on a taxonomy of 10 question types.
Improving the Language Understanding Capabilities of Large Language Models Using Reinforcement Learning (2025.findings-emnlp)

Copied to clipboard

Challenge: Instruction-fine-tuned large language models (LLMs) under 14B parameters underperform on NLU tasks . we explore a framework to improve the NLU capabilities of LLMs .
Approach: They propose to use Proximal Policy Optimization to improve NLU capabilities . they frame NLU as a reinforcement learning environment and optimize for reward signals .
Outcome: The proposed framework outperforms supervised fine-tuning on GLUE and superGLUE tasks.
MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows (2025.findings-naacl)

Copied to clipboard

Challenge: Scientific innovation is driven by detailed workflows, which include critical steps such as contextualizing literature, generating ideas, validating ideas, and planning new research.
Approach: They propose to use large language models to extract five key aspects from scientific publications to optimize scientific workflows.
Outcome: The proposed dataset includes more than 152,000 peer-reviewed publications from 17 leading computer science conferences spanning the past 50 years.
Infusing Sequential Information into Conditional Masked Translation Model with Self-Review Mechanism (2020.coling-main)

Copied to clipboard

Challenge: Existing non-autoregressive models generate target words in parallel, but with a large latency due to the left-to-right dependency.
Approach: They propose to train a conditional masked translation model and refine results within several iterations to remedy a flawed translation by non-autoregressive models.
Outcome: The proposed model outperforms state-of-the-art models by over 1 BLEU while using less training computations.
Semantic Graphs for Generating Deep Questions (2020.acl-main)

Copied to clipboard

Challenge: Existing research has focused on generating factoid questions relevant to one fact obtainable from a single sentence.
Approach: They propose a framework that first constructs a semantic-level graph and then encodes it by introducing an attention-based GGNN.
Outcome: The proposed framework captures the global structure of the document and facilitates reasoning over multiple facts.
Takin-VC: Expressive Zero-Shot Voice Conversion via Adaptive Hybrid Content Encoding and Enhanced Timbre Modeling (2025.acl-long)

Copied to clipboard

Challenge: Expressive zero-shot voice conversion (VC) aims to modify source timbre to match unseen speaker . existing zero- shot VC systems struggle to reproduce paralinguistic information in highly expressive speech .
Approach: They propose a framework for expressive zero-shot voice conversion that uses hybrid content encoding and memory-augmented context-aware timbre modeling.
Outcome: The proposed framework surpasses state-of-the-art VC systems in speech naturalness, speaker similarity, and speaker similarness.
Unlocking Exploration in RLVR: Uncertainty-aware Advantage Shaping for Deeper Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Reinforcement Learning with Verifiable Rewards (RLVR) has shown significant promise for enhancing the reasoning capabilities of large language models (LLMs).
Approach: They propose a model-free method that refines credit assignment by leveraging the model's internal uncertainty signals.
Outcome: Extensive experiments on five mathematical reasoning benchmarks show that the proposed method outperforms strong RLVR baselines on multiple model scales, including 1.5B and 7B.
LEDOM: Reverse Language Model (2026.acl-long)

Copied to clipboard

Challenge: Autoregressive language models are trained exclusively left-to-right, yet they are limited in their ability to factorize text.
Approach: They propose a purely reverse autoregressive language model that factorizes text as a product of left-to-right conditionals.
Outcome: The proposed model can be used to score forward outputs using reverse posterior estimates.
Learning from Few Samples: A Novel Approach for High-Quality Malcode Generation (2025.emnlp-main)

Copied to clipboard

Challenge: Intrusion detection systems (IDS) are limited in labeled samples due to scarcity and lack of diversity in malicious samples.
Approach: They propose a semi-supervised framework that integrates Generative Adversarial Networks with Large Language Models to enhance malicious code generation and SQL Injection detection capabilities.
Outcome: The proposed framework enhances malicious code generation and detection capabilities in few-sample learning scenarios.

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