Papers by Shengyu Li

17 papers
Optimizing Code Retrieval: High-Quality and Scalable Dataset Annotation through Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Existing methods for code retrieval struggle to balance scalability and annotation quality.
Approach: They propose a method that integrates functions called within the repository and information on third-party APIs to enhance the annotation context.
Outcome: The proposed method improves the annotation context by incorporating functions called within the repository and information on third-party API functionalities.
VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Recent research has focused on pushing weight-only quantization to extremely low-bit due to numerical representation limitations.
Approach: They propose a vector-based quantization approach that pushes LLMs to extremely low-bit . they propose scalar-based weight quantization that reduces memory requirements and optimizes storage costs .
Outcome: The proposed method reduces model quantization perplexity by 0.01-0.34 on LLaMA-2, 0.38-0.68 on mistral-7B, 4.41-7.34, on llaMA-3 on QA tasks on average.
Boosting Multi-modal Keyphrase Prediction with Dynamic Chain-of-Thought in Vision-Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Multi-modal keyphrase prediction (MMKP) aims to produce concise, informative phrases that capture the essence of cross-modal inputs.
Approach: They propose to use vision-language models to generate conclusive phrases using multiple modalities of input information.
Outcome: The proposed methods outperform existing methods on absence and unseen scenarios and overestimate model capability due to overlap in training tests.
End-to-End Modeling via Information Tree for One-Shot Natural Language Spatial Video Grounding (2022.acl-long)

Copied to clipboard

Challenge: Existing methods for grounding video frames with dense annotations require enormous amount of human effort.
Approach: They propose to ground natural language in video frames with only one frame labeled . they propose an end-to-end model that eliminates interference of irrelevant frames .
Outcome: The proposed model can ground natural language in all video frames with only one frame labeled . the proposed model eliminates interference of irrelevant frames based on branch search and cropping techniques .
Towards Advanced Mathematical Reasoning for LLMs via First-Order Logic Theorem Proving (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) have shown promising first-order logic (FOL) reasoning capabilities with applications in various areas, but their effectiveness in complex mathematical reasoning involving multi-step FOL deductions remains under-explored.
Approach: They propose a self-adaptive solution that enhances the Diversity and REAsonability of LLMs’ generation strategies by introducing an Axiom-Driven Strategy Diversification mechanism and a Sub-Proposition Error Feedback to help LLM reflect on and correct their proofs.
Outcome: The proposed model improves diversity and REAsonability of LLMs’ generation strategies by introducing an Axiom-Driven Strategy Diversification mechanism and a Sub-Proposition Error Feedback to help LLM reflect on and correct proofs.
MalURLBench: A Benchmark Evaluating Agents’ Vulnerabilities When Processing Web URLs (2026.findings-acl)

Copied to clipboard

Challenge: Existing models struggle to detect elaborately disguised malicious URLs, despite their ability to process malicious URL's.
Approach: They propose a benchmark to evaluate LLMs’ vulnerabilities to malicious URLs and a lightweight defense module to mitigate the vulnerability.
Outcome: The proposed framework analyzes 61,845 attack instances spanning 10 real-world scenarios and 7 categories of real malicious websites.
Weakly-Supervised Spoken Video Grounding via Semantic Interaction Learning (2023.acl-long)

Copied to clipboard

Challenge: Recent work on spoken video grounding challenges extracting semantic information from speech . previous studies focused on textual queries, but recent work focuses on spoken queries .
Approach: They propose a framework for weakly-supervised spoken video grounding to represent cross-modal semantics without expensive temporal annotations.
Outcome: The proposed framework is more efficient than existing methods.
InfiGUIAgent: A Multimodal Generalist GUI Agent with Native Reasoning and Reflection (2026.eacl-long)

Copied to clipboard

Challenge: Existing GUI Agents face challenges in multi-step reasoning and reliance on textual annotations, limiting their effectiveness.
Approach: They propose an MLLM-based GUI Agent with a two-stage supervised fine-tuning pipeline that enhances GUI understanding and grounding.
Outcome: InfiGUIAgent achieves competitive performance on several GUI benchmarks, highlighting the impact of native reasoning skills in enhancing GUI interaction for automation tasks.
Neural Gibbs Sampling for Joint Event Argument Extraction (2020.aacl-main)

Copied to clipboard

Challenge: Existing methods for event argument extraction cannot adequately model the correlation between event arguments and their roles.
Approach: They propose a Bayesian model to jointly extract event arguments using Gibbs sampling . they train two neural networks to model prior distribution and conditional distribution over event arguments .
Outcome: The proposed model can achieve comparable results to existing methods on two widely-used datasets.
ART: rule bAsed futuRe-inference deducTion (2023.emnlp-main)

Copied to clipboard

Challenge: Existing studies focus on language-based premises and deduce valid conclusions from visual observations.
Approach: They propose a rule-based deductive reasoning task that uses video to deduce the correct future event . they use commonsense knowledge to annotate video and a strong baseline to conduct reasoning .
Outcome: Empirical studies validate the rationality of ARTNet in deductive reasoning upon visual observations . ART is a method that rigorously follows a set of explicit constraints to deduce valid conclusions from empirical facts .
PhiloGPT: A Philology-Oriented Large Language Model for Ancient Chinese Manuscripts with Dunhuang as Case Study (2024.emnlp-main)

Copied to clipboard

Challenge: philology requires years of professional training in extensive knowledge memorization and manual textual retrieval.
Approach: They curated the PhiloCorpus-ZH, a rich collec-tion of ancient Chinese texts spanning a millennium with 30 diverse topics, including firsthand folk copies.
Outcome: The PhiloCorpus-ZH corpus facilitated the development of the first LLM tailored for discovering ancient Chinese manuscripts.
Measure Twice, Click Once: Co-evolving Proposer and Visual Critic via Reinforcement Learning for GUI Grounding (2026.acl-long)

Copied to clipboard

Challenge: Graphical User Interface (GUI) grounding requires mapping natural language instructions to precise pixel coordinates due to visually homogeneous elements and dense layouts.
Approach: They propose to replace static consistency strategies with a learnable selection mechanism that selects the optimal target by critiquing its own proposals rendered on the screenshot.
Outcome: The proposed model significantly improves both grounding and critiquing capabilities over 6 benchmarks.
MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference (2025.acl-long)

Copied to clipboard

Challenge: Existing KV cache eviction methods fail to capture modality-specific information, resulting in suboptimal performance.
Approach: They propose a modality-adaptive key-value (KV) cache eviction strategy to enhance the efficiency of multimodal large language models in long-context inference.
Outcome: The proposed method reduces the KV cache memory footprint and model inference latency while maintaining high accuracy across multimodal long-context tasks.
Multi-modal Action Chain Abductive Reasoning (2023.acl-long)

Copied to clipboard

Challenge: Existing models for Abductive Reasoning are limited in their ability to infer the most plausible explanation of incomplete known phenomena.
Approach: They propose a vision-language task that aims to imagine the most plausible event by spatio-temporal grounding in past video and infer the hypothesis of subsequent action chain layer by layer.
Outcome: The proposed model outperforms existing video-language models in terms of effectiveness on the proposed dataset.
OS Agents: A Survey on MLLM-based Agents for Computer, Phone and Browser Use (2025.acl-long)

Copied to clipboard

Challenge: a new generation of (M)LLMs is enabling the creation of superintelligent AI assistants . OS Agents can complete tasks autonomously and have the potential to significantly enhance the lives of billions of users worldwide.
Approach: They propose to build OS Agents that operate within operating systems' GUIs and GUIs . they examine evaluation metrics and benchmarks to identify promising directions .
Outcome: The proposed agents are based on operating systems (OS) and operating systems frameworks.
RaFe: Ranking Feedback Improves Query Rewriting for RAG (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models (LLMs) and Retrieval Augmentation Generation (RAG) techniques have evolved to enhance document retrieval by reformulating queries.
Approach: They propose a framework for training query rewriting models that leverages a reranker framework.
Outcome: The proposed framework provides ranking feedback aligned well with the rewriting objectives without needing signals from annotations and supports both online and offline training models.
AdvancedIF: Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Following (2026.acl-long)

Copied to clipboard

Challenge: Recent advances in large language models (LLMs) have shown impressive performance on a range of tasks, yet advanced instruction following (IF) remains a significant challenge.
Approach: They propose a benchmark that features over 1,600 prompts and expert-curated rubrics that assess LLMs’ ability to follow complex, multi-turn, and system-level instructions.
Outcome: The proposed framework improves instruction-following abilities of large language models, achieving a 6.7% gain on AdvancedIF and strong results on public benchmarks.

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