Papers by Ziyue Li

25 papers
Model Composition for Multimodal Large Language Models (2024.acl-long)

Copied to clipboard

Challenge: Existing methods for creating versatile MLLMs rely on joint training with paired instruction data, which is resource-intensive and challenging to extend to new modalities.
Approach: They propose a new paradigm for multimodal large language models by reusing modality encoders and merging LLM parameters.
Outcome: The proposed model retains the modal understanding capabilities of each original model.
Lil: Less is Less When Applying Post-Training Sparse-Attention Algorithms in Long-Decode Stage (2026.findings-acl)

Copied to clipboard

Challenge: Prior work typically decomposes inference into prefill and decode stages, with the decode stage dominating total latency.
Approach: They propose an algorithm that detects threshold where information loss exceeds information gain during sparse decoding to reduce token consumption by up to 90% and a marginal accuracy degradation of less than 2%.
Outcome: The proposed algorithm reduces token consumption by 90% with a marginal accuracy degradation of less than 2% across reasoning-intensive benchmarks.
Enabling Stroke-Level Structural Analysis of Hieroglyphic Scripts without Language-Specific Priors (2026.findings-acl)

Copied to clipboard

Challenge: Existing structural analysis methods for hieroglyphic scripts are script-specific and labor-intensive.
Approach: They propose a hieroglyphic Stroke Analyzer framework that captures character-internal structures and semantics without handcrafted data.
Outcome: The proposed framework captures character-internal structures and semantics without priors . it can be used to generalize hieroglyphic scripts across languages .
ActiView: Evaluating Active Perception Ability for Multimodal Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Existing benchmarks for evaluating MLLMs have not addressed active perception . a novel benchmark is proposed to evaluate active perception in ML models .
Approach: They propose a benchmark to evaluate active perception in Multimodal Large Language Models . they restrict the perceptual field of a model and require it to actively zoom or shift it .
Outcome: The proposed benchmark focuses on a specialized form of Visual Question Answering (VQA) that eases and quantifies the evaluation yet challenging for existing MLLMs.
Sparser Mixture-of-Adapters with Cross-Layer Generalization (2025.naacl-long)

Copied to clipboard

Challenge: Existing methods for training large language models do not allow sharing adapters across layers . existing methods do not support sharing adapter pools, leading to redundancy and poor generalization .
Approach: They propose a mixture-of-adapter framework that trains a pool of lightweight adapters at each layer and selects the most suitable ones for each input.
Outcome: The proposed framework reduces active adapters by over 85% while boosting task accuracy.
Perspective Transition of Large Language Models for Solving Subjective Tasks (2025.findings-acl)

Copied to clipboard

Challenge: Large language models (LLMs) have revolutionized the field of natural language processing . performance of LLMs on subjective tasks is limited, authors say .
Approach: They propose a method that allows LLMs to select between direct, role, and third-person perspectives for best way to solve corresponding subjective problem.
Outcome: The proposed method outperforms widely used single fixed perspective based methods on 12 subjective tasks.
Browse and Concentrate: Comprehending Multimodal Content via Prior-LLM Context Fusion (2024.acl-long)

Copied to clipboard

Challenge: Multimodal Large Language Models (MLLMs) lack understanding of multi-image and interleaved inputs due to the visual features encoded by frozen encoders before being fed into the LLM backbone.
Approach: They propose a two phase paradigm to enable in-depth multimodal context fusion prior to feeding the features into LLMs.
Outcome: The proposed paradigm boosts the performance on 7 multi-image scenarios, contributing to increments on average accuracy by 2.13% and 7.60% against strong MLLMs baselines with 3B and 11B LLMs, respectively.
TCSinger: Zero-Shot Singing Voice Synthesis with Style Transfer and Multi-Level Style Control (2024.emnlp-main)

Copied to clipboard

Challenge: Existing models fail to generate singing voices rich in stylistic nuances for unseen singers due to multifaceted nature of singing styles.
Approach: They propose a zero-shot SVS model for style transfer across cross-lingual speech and singing styles and multi-level style control.
Outcome: Experimental results show that TCSinger outperforms baseline models in synthesis quality, singer similarity, and style controllability.
TreeReview: A Dynamic Tree of Questions Framework for Deep and Efficient LLM-based Scientific Peer Review (2025.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) have shown significant potential in assisting peer review, but current methods struggle to generate thorough and insightful reviews while maintaining efficiency.
Approach: They propose a framework that models paper review as a hierarchical and bidirectional question-answering process.
Outcome: The proposed framework outperforms baselines on full review generation and actionable feedback comments generation tasks while reducing LLM token usage by up to 80% compared to computationally intensive approaches.
Diving into Mitigating Hallucinations from a Vision Perspective for Large Vision-Language Models (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing benchmarks focus on coarse-grained hallucination detection and fail to capture hallucinics . vision encoders exhibit unique hallucinian characteristics, but suboptimal of simple feature fusion.
Approach: They propose a visual encoder that employs different training paradigms to instill inductive biases in visual encoded models.
Outcome: The proposed system reduces hallucinations and improves model performance.
Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios (2026.findings-acl)

Copied to clipboard

Challenge: Existing evaluation benchmarks for long-form speech are limited to limited domains, creating a significant gap with the diverse downstream applications.
Approach: They propose a benchmark that decomposes "long-form speech quality" into specific, disentangled dimensions.
Outcome: The proposed benchmark decomposes “long-form speech quality” into specific, disentangled dimensions.
CAMIEval: Enhancing NLG Evaluation through Multidimensional Comparative Instruction-Following Analysis (2025.naacl-long)

Copied to clipboard

Challenge: Evaluating the quality of texts generated by language models has always been a challenging task in natural language processing (NLP).
Approach: They propose a multidimensional comparative evaluation method based on instruction-following that combines relevance, factuality, and adherence with a concrete Chain-of-Thoughts process to enhance the accuracy of evaluations.
Outcome: The proposed method outperforms existing methods in correlation with human evaluations on two NLG evaluation benchmarks.
How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients (2026.acl-long)

Copied to clipboard

Challenge: Spectral properties of low/high-quality instruction and reasoning data are used to explain finetuning dynamics in large language models.
Approach: They propose to analyze layer-wise gradients induced by low/high-quality instruction and reasoning data for LLM post-training.
Outcome: The results show that higher-quality data are associated with lower nuclear norms and higher effective ranks.
TPTU-v2: Boosting Task Planning and Tool Usage of Large Language Model-based Agents in Real-world Industry Systems (2024.emnlp-industry)

Copied to clipboard

Challenge: Large language models have demonstrated proficiency in addressing tasks that necessitate a combination of task planning and the usage of external tools.
Approach: They propose a framework to enhance the task planning and tool usage abilities of LLMs in industrial systems.
Outcome: The proposed framework enhances the task planning and tool usage abilities of LLM-based agents in industrial systems.
MLLM-Bench: Evaluating Multimodal LLMs with Per-sample Criteria (2025.naacl-long)

Copied to clipboard

Challenge: Existing evaluation methodologies for multimodal large language models are limited in evaluating objective queries without considering real-world user experiences.
Approach: They propose to evaluate multimodal large language models with per-sample criteria using potent MLLM as the judge.
Outcome: The proposed evaluation paradigm shows that it can be used to evaluate multimodal large language models with per-sample criteria.
Rhythm Controllable and Efficient Zero-Shot Voice Conversion via Shortcut Flow Matching (2025.acl-long)

Copied to clipboard

Challenge: Existing methods focus on disentangling speakers and content, while others focus on preserving the source's prosody.
Approach: They propose a rhythm-controllable and efficient zero-shot voice conversion model that transforms the source speaker’s timbre into an unseen one while retaining speech content.
Outcome: The proposed model adapts the linguistic content duration to the desired speaking style, facilitating the transfer of the target speaker’s rhythm.
CODIS: Benchmarking Context-dependent Visual Comprehension for Multimodal Large Language Models (2024.acl-long)

Copied to clipboard

Challenge: Multimodal large language models have demonstrated promising results in a variety of tasks that combine vision and language.
Approach: They propose a benchmark to assess the ability of models to use contextual information in free-form text to enhance visual comprehension.
Outcome: The proposed model fails to extract and utilize contextual information to improve understanding of images.
FiNE: Filtering and Improving Noisy Data Elaborately with Large Language Models (2025.naacl-long)

Copied to clipboard

Challenge: Currently, there are two mainstream methods for improving data integrity: data filtering and data augmentation.
Approach: They propose a method to improve data integrity by combining data filtering and data augmentation with LLMs.
Outcome: The proposed method surpasses the open-source chat version on HalluQA by 8.45 on the open source version.
MUCAR: Benchmarking Multilingual Cross-Modal Ambiguity Resolution for Multimodal Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing multimodal benchmarks overlook linguistic and visual ambiguities, authors say . ambiguity resolution between modalities is lacking in multimodal large language models .
Approach: They propose a benchmark to evaluate multimodal ambiguity resolution across multilingual and cross-modal scenarios.
Outcome: a new benchmark evaluates multimodal ambiguity resolution across multilingual and cross-modal scenarios . the benchmark shows that MLLMs can resolve ambiguities in image-text alignment . however, existing benchmarks often overlook linguistic and visual ambiguties .
MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for MLLMs struggle with fine-grained temporal reasoning . despite advances in video understanding, current methods struggle with time-sensitive tasks .
Approach: They propose a time-stamp-aware multi-segment grounding method that enhances temporal understanding by introducing timestamps.
Outcome: The proposed method outperforms existing methods on time-sensitive tasks and generalizes well across diverse temporal understanding scenarios.
Versatile Framework for Song Generation with Prompt-based Control (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for song generation fail to generate vocals with prompt-based control and proper alignment.
Approach: VersBand is a multi-task song generation framework for synthesizing high-quality songs with prompt-based control.
Outcome: Experimental results show that VersBand performs better than baseline models across multiple song generation tasks.
BrainLoc: Brain Signal-Based Object Detection with Multi-modal Alignment (2025.findings-emnlp)

Copied to clipboard

Challenge: BrainLoc is a lightweight object detection model guided by fMRI signals.
Approach: They propose a brain-based object detection model guided by fMRI signals . they employ a multi-modal alignment strategy that enhances fmr feature extraction .
Outcome: The proposed model improves fMRI-based object detection accuracy and convenience.
Filling the Image Information Gap for VQA: Prompting Large Language Models to Proactively Ask Questions (2023.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) demonstrate impressive reasoning ability and the maintenance of world knowledge in natural language tasks.
Approach: They propose a framework that enables LLMs to ask relevant questions to uncover more details in the image, along with filters for refining the generated information.
Outcome: The proposed framework boosts the performance of baseline methods by 2.15% on OK-VQA and achieves consistent improvements across different LLMs.
GuiLoMo: Allocating Experts and Ranks for LoRA-MoE via Bilevel Optimization with GuidedSelection Vectors (2025.findings-emnlp)

Copied to clipboard

Challenge: Low-Rank Adaptation (LoRA) methods are efficient for a large language model with reduced computational costs.
Approach: They propose a layer-wise expert numbers and ranks allocation strategy with GuidedSelection Vectors.
Outcome: The proposed method achieves superior or comparable performance to all baselines on three backbone models.
DongbaMIE: A Multimodal Information Extraction Dataset for Evaluating Semantic Understanding of Dongba Pictograms (2025.findings-emnlp)

Copied to clipboard

Challenge: Dongba pictographic is the only pictograph script still in use in the world.
Approach: DongbaMIE is the first dataset focusing on multimodal information extraction of Dongbe pictographs.
Outcome: The dataset contains 23,530 sentence-level and 2,539 paragraph-level high-quality text-image pairs.

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