Papers by Ma Jun

39 papers
Static Models, Dynamic World: A Unified Perspective on Temporal Perception in Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Large language models are trained on static corpora but deployed in a dynamic world . a foundational tension remains between time and the ability to understand it .
Approach: They formalize temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers.
Outcome: The proposed framework formalizes temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers . the framework induces four temporal information regimes corresponding to internal reasoning, answer recency, premise anchoring, and genuine world indeterminacy .
Unified Thinker: A General Reasoning Core for Image Generation (2026.acl-long)

Copied to clipboard

Challenge: generative models struggle with logic-intensive instruction following, exposing a persistent reasoning–execution gap.
Approach: They propose a task-agnostic reasoning architecture for general image generation . they propose pixel-level feedback to ground the Thinker's policy in pixel feedback .
Outcome: The proposed system significantly improves image reasoning and generation quality.
MenatQA: A New Dataset for Testing the Temporal Comprehension and Reasoning Abilities of Large Language Models (2023.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) have shown nearly saturated performance on many NLP tasks.
Approach: They construct multiple sensitive factors time QA which encompasses three temporal factors . they test current mainstream LLMs with different parameter sizes .
Outcome: The proposed model incorporates three temporal factors with 2,853 samples . the results show that LLMs fall behind smaller models on these factors .
Few-shot Named Entity Recognition with Entity-level Prototypical Network Enhanced by Dispersedly Distributed Prototypes (2022.coling-1)

Copied to clipboard

Challenge: Existing prototypical networks for named entity recognition suffer from label dependency and tightly distributed prototypes, thus causing misclassifications.
Approach: They propose an Entity-level Prototypical Network enhanced by dispersedly distributed prototypes to build entity-level prototypes and distribute them dispersionally.
Outcome: The proposed system outperforms the previous models on two evaluation tasks and the Few-NERD settings in terms of overall performance.
GeoLaux: A Benchmark for Evaluating MLLMs’ Geometry Performance on Long-Step Problems Requiring Auxiliary Lines (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks for Geometry problem solving lack fine-grained evaluation for long-step problems necessitating auxiliary line construction.
Approach: They present a fine-grained annotated dataset with long-step reasoning and auxiliary line construction that provides a detailed evaluation of 23 leading MLLMs.
Outcome: The proposed model performs significantly worse on long-step problems than short-step ones, with 18 models showing a performance drop of over 50%.
DB-Explore: Automated Database Exploration and Instruction Synthesis for Text-to-SQL (2025.findings-emnlp)

Copied to clipboard

Challenge: Recent text-to-SQL systems that use large language models struggle with complex database structures and domain-specific queries.
Approach: a framework that aligns large language models with database knowledge is proposed . DB-Explore constructs database graphs to capture complex relational schemas .
Outcome: a new framework outperforms existing text-to-SQL systems by outperforming existing systems.
LaMP-Val: Large Language Models Empower Personalized Valuation in Auction (2025.findings-emnlp)

Copied to clipboard

Challenge: Currently, most research focuses on the bidding algorithms used within auction mechanisms.
Approach: They propose a personalized valuation framework that integrates Large Language Models to incorporate personalized semantic preference into users valuation process.
Outcome: The proposed framework incorporates Large Language Models to incorporate personalized semantic preference into users valuation process.
Title2Event: Benchmarking Open Event Extraction with a Large-scale Chinese Title Dataset (2022.emnlp-main)

Copied to clipboard

Challenge: Existing EE datasets define fixed event types and design specific schemas for each of them, failing to cover diverse events emerging from the online text.
Approach: They propose to use a sentence-level dataset to benchmark Open Event Extraction without restricting event types.
Outcome: The proposed dataset contains more than 42,000 news titles in 34 topics collected from Chinese web pages.
Pause or Fabricate? Training Language Models for Grounded Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Large language models implicitly fabricate information when inputs are incomplete, causing confidence but unreliable conclusions.
Approach: They propose a framework for grounded reasoning under incomplete information that decomposes reasoning into two stages . they propose stage-specific rewards to penalize hallucinations, enabling models to detect gaps, stop proactively, and resume reasoning after clarification.
Outcome: The proposed framework improves premise detection and task success by 30% . it also reduces average response length by over 20% .
Robust Preference Optimization via Dynamic Target Margins (2025.findings-acl)

Copied to clipboard

Challenge: Direct Preference Optimization (DPO) is an efficient method for ensuring safety and reliability in practical applications.
Approach: They propose a dynamic target margin preference optimization algorithm that adjusts reward margins at the pairwise level.
Outcome: The proposed method achieves an average 4.4% improvement over baselines, setting new benchmarks for state-of-the-art performance.
End-to-End Conversational Search for Online Shopping with Utterance Transfer (2021.emnlp-main)

Copied to clipboard

Challenge: a new study proposes a conversational search system that integrates product attributes and dialog with search . but it faces two real world challenges: imperfect product schema/knowledge and lack of training dialog data .
Approach: They propose an end-to-end conversational search system that integrates search with text . they propose an utterance transfer approach that generates dialogue utterations from other domains .
Outcome: The proposed system outperforms the best tested baseline in a conversational search dataset for online shopping.
Multi-Hop Knowledge Editing via Critic-Guided Multi-Agent Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing knowledge editing methods rely on unidirectional, feed-forward pipelines . a minor retrieval error or logical mismatch at an early hop can become a silent failure .
Approach: They propose a framework for closed-loop post-edit reasoning that uses a Critic agent to verify coherence and step-wise correctness.
Outcome: Experiments on MQuAKE-2002 and MQuADE-hard show that CARE effectively mitigates error propagation . a minor retrieval error or logical mismatch at an early hop can become a silent failure .
TXtract: Taxonomy-Aware Knowledge Extraction for Thousands of Product Categories (2020.acl-main)

Copied to clipboard

Challenge: State-of-the-art methods for knowledge extraction are designed for a single category of product, but do not apply to real-life e-Commerce scenarios.
Approach: They propose a taxonomy-aware knowledge extraction model that applies to thousands of categories organized in a hierarchical taxonomies.
Outcome: The proposed model outperforms state-of-the-art methods on 4,000 categories in F1 and 15% across all categories.
UniTabNet: Bridging Vision and Language Models for Enhanced Table Structure Recognition (2024.findings-emnlp)

Copied to clipboard

Challenge: Table structure recognition technology is a critical tool for processing and analyzing large volumes of tabular data.
Approach: They propose a framework for table structure parsing based on the image-to-text model and a vision guider to refine the model’s capability to understand textual semantics in table images.
Outcome: The proposed framework improves on a dataset of PubTabNet, PubTables1M, WTW, and iFLYTAB and will be made publicly available.
Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition (2026.findings-acl)

Copied to clipboard

Challenge: Existing approaches to GMNER use MLLMs as auxiliary tools, causing cumulative error propagation and a lack of rigorous cross-modal verification.
Approach: They propose a model that enforces structured cross-modal reasoning through Multi-style Reasoning Schema Injection and Constraint-guided Verifiable Optimization.
Outcome: The proposed model enforces structured cross-modal reasoning through multi-style Reasoning Schema Injection and Constraint-guided Verifiable Optimization.
Genius: A Generalizable and Purely Unsupervised Self-Training Framework For Advanced Reasoning (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for enhancing LLM reasoning rely on supervisory signals . current methods rely heavily on outcome supervision and auxiliary reward models .
Approach: They propose a gen-eralizable and purely unsupervised self-training framework to enhance LLM reasoning without supervision.
Outcome: The proposed framework improves LLM reasoning without supervision without external supervision.
PivotFEC: Enhancing Few-shot Factual Error Correction with a Pivot Task Approach using Large Language Models (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for Factual Error Correction (FEC) use mask-then-correct paradigms . however, the lack of datasets containing false claims has impeded progress .
Approach: They propose a method that enhances few-shot FEC with a pivot task approach using large language models.
Outcome: The proposed method outperforms its few-shot counterpart by 7.9 points in SARI . it improves widely-adopted SARI metrics by 11.3 compared to the best-performing methods .
LinkNBed: Multi-Graph Representation Learning with Entity Linkage (P18-1)

Copied to clipboard

Challenge: Knowledge graphs have emerged as an important model for studying complex multi-relational data.
Approach: They propose a deep relational learning framework that learns entity and relationship representations across multiple graphs.
Outcome: The proposed framework improves on the state-of-the-art relational learning approaches and identifies entity linkage across graphs.
Agent-Dice: Disentangling Knowledge Updates via Geometric Consensus for Agent Continual Learning (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Model (LLM)-based agents extend the utility of LLMs by interacting with dynamic environments.
Approach: They propose a parameter fusion framework based on directional consensus evaluation that disentangles knowledge updates through a two-stage process.
Outcome: The proposed framework disentangles knowledge updates through a two-stage process with minimal computational overhead and parameter updates.
MobiLoRA: Accelerating LoRA-based LLM Inference on Mobile Devices via Context-aware KV Cache Optimization (2025.acl-long)

Copied to clipboard

Challenge: MobiLoRA focuses on optimizing the key-value (KV) caches due to the limited computing and memory resources of mobile devices.
Approach: They propose to optimize the key-value caches due to limited computing resources . they propose similarity-aware delta encoding for semantic-level contexts .
Outcome: The proposed model accelerates LoRA-based LLM inference by 57.6% on mobile devices.
Fine-Mem: Fine-Grained Feedback Alignment for Long-Horizon Memory Management (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to memory management rely on final task performance as the primary reward, resulting in severe reward sparsity and ineffective credit assignment.
Approach: They propose a framework for fine-grained feedback alignment using a Chunk-level step reward and Evidence-Anchored Reward Attribution to redistribute global rewards based on memory items utilized as evidence in reasoning.
Outcome: The proposed framework outperforms baselines and supports generalization across different model configurations and backbones.
Interleaved Tool-Call Reasoning for Protein Function Understanding (2026.acl-long)

Copied to clipboard

Challenge: Recent advances in large language models have highlighted the effectiveness of chain-of-thought reasoning in symbolic domains such as mathematics and programming.
Approach: They propose a tool-augmented protein reasoning agent that unifies problem decomposition, tool invocation, and grounded answer generation.
Outcome: The proposed protein function understanding agent outperforms text-only reasoning models with an average performance improvement of 103%.
Making MLLMs Blind: Adversarial Smuggling Attacks in MLLM Content Moderation (2026.findings-acl)

Copied to clipboard

Challenge: Multimodal Large Language Models (MLLMs) are increasingly being deployed as content moderators . however, they exploit the Human-AI capability gap and create adversarial environments . smuggling attacks exploit the human-AI gap and exploit the vulnerability .
Approach: They construct a benchmark to evaluate the vulnerability of MLLMs as content moderators . they identify three root causes: limited capabilities of vision encoders, robustness gap in OCR .
Outcome: The proposed model exploits the Human-AI capability gap and is vulnerable to smuggling attacks.
UI-Copilot: Advancing Long-Horizon GUI Automation via Tool-Integrated Policy Optimization (2026.acl-long)

Copied to clipboard

Challenge: Experimental results show that UI-Copilot-7B achieves state-of-the-art performance on challenging MemGUI-Bench, outperforming strong 7B-scale GUI agents such as GUI-Owl-7B and UITARS-1.5-7B.
Approach: They propose a collaborative framework where the GUI agent focuses on task execution while a lightweight copilot provides on-demand assistance for memory retrieval and numerical computation.
Outcome: The proposed framework outperforms GUI-Owl-7B and UI-TARS-1.5-7B on MemGUI-Bench and delivers 17.1% improvement on AndroidWorld over the base Qwen model.
CIF-PT: Bridging Speech and Text Representations for Spoken Language Understanding via Continuous Integrate-and-Fire Pre-Training (2023.findings-acl)

Copied to clipboard

Challenge: Speech-to-text training and language model distillation are used to bridge the representations between speech and text.
Approach: They propose a pre-training paradigm that integrates speech and text into a single frame-to-token alignment.
Outcome: The proposed paradigm outperforms the state-of-the-art model on intent classification and slot filling tasks.
Efficient Paths and Dense Rewards: Probabilistic Flow Reasoning for Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to mitigate inference inefficiency and optimization difficulty are fragmented and constrained by inherent trade-offs.
Approach: They propose a framework that reconceptualizes discrete reasoning steps as a continuous probabilistic flow, quantifying the contribution of each step toward the ground-truth answer.
Outcome: The proposed framework achieves a superior balance between inference efficiency and reasoning performance on challenging benchmarks.
Subgraph-Guided Executable Logical Form Generation for Knowledge Base Question Answering (2026.findings-acl)

Copied to clipboard

Challenge: Existing retrieval-augmented approaches focus on ignoring the structural information of the Knowledge Base (KB) and the question.
Approach: They propose a structure-aware subgraph retrieval stage that ranks candidate subgraphs by aligning them with the question’s structure, along with semantic relevance.
Outcome: Experiments on GrailQA, WebQSP, and GraphQuestions show that the proposed framework achieves state-of-the-art performance.
LLaSE-G1: Incentivizing Generalization Capability for LLaMA-based Speech Enhancement (2025.acl-long)

Copied to clipboard

Challenge: Recent advances in language models have demonstrated strong capabilities in semantic understanding and contextual modeling.
Approach: They propose a LLaMA-based language model that incentivizes generalization capabilities for speech enhancement.
Outcome: The proposed language model outperforms prior task-specific discriminative and generative models in acoustic enhancement tasks.
3AM: An Ambiguity-Aware Multi-Modal Machine Translation Dataset (2024.lrec-main)

Copied to clipboard

Challenge: Existing studies have shown that visual information in existing MMT datasets is insufficient, causing models to disregard it and overestimate their capabilities.
Approach: They propose to use 3AM to create an ambiguity-aware multimodal machine translation dataset.
Outcome: The proposed dataset includes more ambiguity and a greater variety of captions and images than other MMT datasets.
Train in Vain: Functionality-Preserving Poisoning to Prevent Unauthorized Use of Code Datasets (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for dataset poisoning require full-dataset poison, which breaks code compilability.
Approach: They propose a functionality-preserving poisoning approach that injects short, compilable weak-use fragments into executed code paths.
Outcome: The proposed method contaminates 10% of the dataset while maintaining 100% compilability and functional correctness.
PILE: Pairwise Iterative Logits Ensemble for Multi-Teacher Labeled Distillation (2022.emnlp-industry)

Copied to clipboard

Challenge: Pre-trained language models have been a key part of ranking systems . knowledge distillation is widely used to maintain high performance while keeping efficient computations.
Approach: They propose an algorithm to combine knowledge from multi-teachers and label information to achieve competitive performance in offline and online experiments.
Outcome: The proposed method has been deployed in a real-world commercial search system.
A Wolf in Sheep’s Clothing: Generalized Nested Jailbreak Prompts can Fool Large Language Models Easily (2024.naacl-long)

Copied to clipboard

Challenge: Existing methods for generating 'jailbreaks' suffer from manual design or require optimization on other white-box models, which compromises either generalization or efficiency.
Approach: They propose a framework that leverages LLMs to generate effective jailbreak prompts and a generalized framework that can be used to generate prompts.
Outcome: The proposed framework improves the attack success rate while reducing the time cost compared to baselines.
UEGP: Unified Expert-Guided Pre-training for Knowledge Rekindle (2024.findings-naacl)

Copied to clipboard

Challenge: Existing paradigms for pre-training and fine-tuning have limitations . knowledge rekindle aims to break through performance upper bounds of experts without introducing additional annotated data.
Approach: They propose a new paradigm for pre-training and fine-tuning that aims to re-incorporate the fine- tuned expert model into the training cycle and break through performance upper bounds of experts.
Outcome: The proposed model breaks through performance upper bounds of experts without additional annotated data.
Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods to generate human-aligned content with a “jailbreak prompt” are inefficient and repetitive, causing inefficiency and a lack of experience.
Approach: They propose a framework that integrates past attack experiences to aid current jailbreak attempts.
Outcome: The proposed framework improves both attack effectiveness and efficiency compared to the current black-box jailbreak method.
𝜙-Decoding: Adaptive Foresight Sampling for Balanced Inference-Time Exploration and Exploitation (2025.acl-long)

Copied to clipboard

Challenge: Existing inference-time optimization strategies address the shortsightedness of auto-regressive generation, but the vast search space leads to excessive exploration and insufficient exploitation.
Approach: They propose a decoding strategy that approximates two distributions via foresight and clustering to provide an efficient estimation of step value.
Outcome: The proposed decoding strategy outperforms strong baselines in performance and efficiency.
SepSeq: A Training-Free Framework for Long Numerical Sequence Processing in LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Existing large-scale large-context models suffer from performance degradation when processing long numerical sequences.
Approach: They propose a framework to mitigate attention dispersion by strategically inserting separator tokens into the model to recalibrat attention to local segments while preserving global context.
Outcome: The proposed framework improves accuracy and reduces inference token consumption by 16.4% on 9 widely-adopted LLMs.
TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing (2021.acl-demo)

Copied to clipboard

Challenge: Existing approaches to textual robustness evaluation focus on slightly modifying the input data, which maintains the original meaning and results in a different prediction.
Approach: They propose a multilingual robustness evaluation toolkit for NLP that integrates universal text transformations, task-specific transformations and adversarial attack.
Outcome: The toolkit includes universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analyses.
Span-based Joint Entity and Relation Extraction with Attention-based Span-specific and Contextual Semantic Representations (2020.coling-main)

Copied to clipboard

Challenge: Existing methods treat each span token equally important, ignoring significant features.
Approach: They propose a span-based joint extraction framework with attention-based semantic representations that utilizes span-specific and contextual representations.
Outcome: The proposed model outperforms existing models on ACE2005, CoNLL2004 and ADE.
CompKBQA: Component-wise Task Decomposition for Knowledge Base Question Answering (2025.emnlp-main)

Copied to clipboard

Challenge: Existing knowledge base question answering methods struggle with complex queries.
Approach: They propose a framework that optimizes the process of fine-tuning a LLM for generating logical forms by enabling it to learn relevant sub-tasks like skeleton generation, topic entity generation, and relevant relations generation.
Outcome: The proposed framework achieves state-of-the-art on two benchmark KBQA datasets, WebQSP and CWQ.

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