Papers by Shiyu Wang

34 papers
Lookahead Q-Cache: Achieving More Consistent KV Cache Eviction via Pseudo Query (2025.emnlp-main)

Copied to clipboard

Challenge: Existing KV cache eviction methods prune tokens using prefilling-stage attention scores, causing inconsistency with actual inference queries.
Approach: They propose a lookahead q-cache framework that generates low-cost pseudo lookaheaded queries to better approximate the true decoding-stage queries.
Outcome: The proposed framework outperforms existing methods on LongBench and Needle-in-a-Haystack benchmarks and can be flexibly combined to yield further improvements.
EvoSpark: Endogenous Interactive Agent Societies for Unified Long-Horizon Narrative Evolution (2026.acl-long)

Copied to clipboard

Challenge: Existing systems suffer from social memory stacking and narrative-spatial dissonance . long-horizon narratives suffer from conflicting relational states without resolution .
Approach: They propose a framework to sustain logically coherent long-horizon narratives within endogenous interactive agent societies.
Outcome: Experiments show that the framework outperforms baselines across paradigms.
Self-Supervised Learning for Contextualized Extractive Summarization (P19-1)

Copied to clipboard

Challenge: Existing models for extractive summarization are usually trained from scratch with a cross-entropy loss . previous work builds an end-to-end system to learn to choose sentences without explicitly modeling document context .
Approach: They propose three auxiliary pre-training tasks that learn to capture the document context in a self-supervised fashion.
Outcome: The proposed models outperform existing models on a CNN/DM dataset.
Simple yet Effective Bridge Reasoning for Open-Domain Multi-Hop Question Answering (D19-58)

Copied to clipboard

Challenge: Existing work on open-domain multi-hop question answering relies on off-the-shelf information retrieval techniques to retrieve answer passages.
Approach: They propose a new subproblem for open-domain multi-hop question answering . they aim to recognize the anchor from a set of start passages with a reading comprehension model .
Outcome: The proposed method significantly improves the baseline method on the open-domain hotpotQA benchmark.
Extracting Multiple-Relations in One-Pass with Pre-Trained Transformers (P19-1)

Copied to clipboard

Challenge: Existing approaches to extract multiple relations from a paragraph require multiple passes over the paragraph.
Approach: They propose a method to extract multiple relations from a paragraph by encoding the paragraph only once.
Outcome: The proposed approach can perform state-of-the-art on the benchmark ACE 2005.
ActionStudio: A Lightweight Framework for Data and Training of Large Action Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing infrastructure for efficient agentic data processing and model training remains underdeveloped.
Approach: They propose a lightweight and extensible data and training framework for large action models . they propose to unify diverse agent trajectories using Unified Format 2.0 .
Outcome: The proposed framework shows 9 higher throughput than existing frameworks and performs well across public and realistic agent benchmarks.
Out-of-Domain Detection for Low-Resource Text Classification Tasks (D19-1)

Copied to clipboard

Challenge: Existing methods for OOD detection and ID classification tasks require massive amounts of ID labeled data and no OOD labeles.
Approach: They propose to use OOD-resistant Prototypical Network to detect OOD cases with limited in-domain (ID) training data to solve this task.
Outcome: The proposed solution outperforms state-of-the-art methods in zero-shot OOD detection task while maintaining a competitive performance on ID classification task.
Putting Captions to the Test: Evaluating Video Caption Quality through Multiple-Choice Question Answering (2026.acl-long)

Copied to clipboard

Challenge: Existing metrics for video captioning are based on text-based comparisons with ground-truth references.
Approach: They propose a reference-free benchmark that assesses video captions based on their utility . they will release the benchmark to facilitate reproducible research .
Outcome: The proposed benchmark improves on human-verified, fine-grained questions . it correlates significantly better with human judgments than existing metrics .
A Learning Rate Path Switching Training Paradigm for Version Updates of Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Version updates are an indispensable requirement for Large Language Models . a large learning rate in the first stage and a complete learning decay process are crucial for version updates of LLMs.
Approach: They propose a learning rate path switching training paradigm for version updates of Large Language Models.
Outcome: The proposed paradigm reduces training cost to 58% when training four versions of LLMs compared to PTFS and CPT .
One-Shot Relational Learning for Knowledge Graphs (D18-1)

Copied to clipboard

Challenge: Existing studies on knowledge graph completion require a large number of positive examples for each relation, but long-tail relations are more common in KGs and those newly added relations do not have many known triples for training.
Approach: They propose a one-shot relational learning framework that utilizes the knowledge distilled by embedding models and learns a matching metric by considering both the learned embeddments and one-hop graph structures.
Outcome: The proposed framework improves on existing embedding models and eliminates the need for retraining when dealing with newly added relations.
Query and Extract: Refining Event Extraction as Type-oriented Binary Decoding (2022.findings-acl)

Copied to clipboard

Challenge: Existing approaches to event extraction are limited to a set of pre-defined types.
Approach: They propose a natural language query framework that uses event types and argument roles to extract candidate triggers and arguments from input text.
Outcome: The proposed framework outperforms existing methods on zero-shot event extraction.
ADEPT-SQL: A High-performance Text-to-SQL Application for Real-World Enterprise-Level Databases (2025.acl-demo)

Copied to clipboard

Challenge: et al., 2017) address domain-specific knowledge barriers, schemas complexity, and computational costs of large LLMs.
Approach: They propose a domain-adapted Text2SQL system that addresses critical deployment challenges in professional fields.
Outcome: The proposed system achieves 97% execution accuracy on real-world databases . it is faster than existing systems and has a higher performance than existing ones.
Multi-Level Cross-Modal Alignment for Speech Relation Extraction (2024.emnlp-main)

Copied to clipboard

Challenge: Existing studies use synthetic speech to train and evaluate SpeechRE models, hindering their development . modality gap issue limits performance of existing models, limiting future researches .
Approach: They propose to use speech data to train and evaluate SpeechRE models by using real speech . they propose to train a cross-modal alignment model to bridge the modality gap .
Outcome: The proposed model can train to bridge the modality gap between speech encoder and text decoder . the proposed model is based on two real SpeechRE datasets .
Sentence Embedding Alignment for Lifelong Relation Extraction (N19-1)

Copied to clipboard

Challenge: Existing approaches to relation extraction require a fixed set of relations . Existing methods assume a closed set of relationships and perform once-and-for-all training on a set of datasets.
Approach: They propose to improve the stochastic gradient methods with a replay memory to alleviate the forgetting problem by anchoring the sentence embedding space.
Outcome: The proposed method outperforms state-of-the-art methods on multiple benchmarks.
Improving Question Answering over Incomplete KBs with Knowledge-Aware Reader (P19-1)

Copied to clipboard

Challenge: Existing models that use incomplete knowledge bases and text data to answer open-domain questions are insufficient to cover full evidence.
Approach: They propose a model which learns to aggregate answer evidence from incomplete knowledge bases and text snippets.
Outcome: The proposed model improves on the widely-used KBQA benchmark WebQSP across settings with different extents of incompleteness.
Seer Self-Consistency: Advance Budget Estimation for Adaptive Test-Time Scaling (2026.findings-acl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated exceptional performance in reasoning tasks, particularly in mathematics.
Approach: They propose a dynamic self-consistency framework that integrates System 1 and System 2 reasoning to improve token efficiency and latency.
Outcome: The proposed method outperforms existing methods, achieving up to 47% reduction in token consumption and 43% reduction in inference latency without significant performance loss.
Explicit Alignment and Many-to-many Entailment Based Reasoning for Conversational Machine Reading (2023.findings-emnlp)

Copied to clipboard

Challenge: Recent research has explored how to improve the abilities of decision-making and question generation.
Approach: They propose a pipeline framework that aligns the document and user-provided information in an explicit way, makes decisions using a lightweight many-to-many entailment reasoning module and generates follow-up questions based on the document.
Outcome: The proposed framework achieves state-of-the-art in micro-accuracy and ranks the first place on the public leaderboard of the CMR benchmark dataset ShARC.
AdaReTaKe: Adaptive Redundancy Reduction to Perceive Longer for Video-language Understanding (2025.findings-acl)

Copied to clipboard

Challenge: Multimodal Large Language Models (MLLMs) are limited by context length when processing long videos.
Approach: They propose a training-free method that flexibly reduces redundancy by allocating compression ratios among time and model layers with theoretical guarantees.
Outcome: Experiments on videoMME, MLVU, LongVideoBench, and LVBench show that AdaRETAKE outperforms existing methods by 2.3% and 2.8% for 7B and 72B models.
Enhancing LLM-Based Persuasion Simulations with Cultural and Speaker-Specific Information (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to persuasive dialogue generation suffer from stance oscillation and low informativeness.
Approach: They propose reinforced instructional prompting, a method that ensures speaker characteristics consistently guide all stages of dialogue generation.
Outcome: The proposed method ensures speaker characteristics guide all stages of dialogue generation and aligns language use with speakers’ native languages to better capture cultural nuances.
Token-Budget-Aware LLM Reasoning (2025.findings-acl)

Copied to clipboard

Challenge: Existing methods to enhance reasoning capabilities of large language models incur significant overhead in token usage, leading to increased costs.
Approach: They propose a token-budget-aware LLM reasoning framework that adjusts the number of reasoning tokens based on the reasoning complexity of each problem.
Outcome: The proposed method reduces token costs in CoT reasoning with only a slight performance reduction.
VCORE: Variance-Controlled Optimization-based Reweighting for Chain-of-Thought Supervision (2026.acl-long)

Copied to clipboard

Challenge: Empirical evaluations demonstrate that VCORE achieves the strongest overall average performance, with especially clear gains on lower-capacity models.
Approach: They propose a framework that reformulates supervision as a constrained optimization problem.
Outcome: Empirical evaluations show that VCORE achieves the strongest overall average performance, with especially clear gains on lower-capacity models.
BAPO: Boundary-Aware Policy Optimization for Reliable Agentic Search (2026.findings-acl)

Copied to clipboard

Challenge: Existing RL-based agentic search models fail to recognize reasoning boundaries and rarely admit "I DON'T KNOW" lack of reliability leads to plausible but unreliable answers, introducing significant risks .
Approach: They propose a framework to cultivate reliable boundary awareness without compromising accuracy.
Outcome: Experiments show that the proposed framework improves the reliability of agentic search models.
STReasoner: Empowering LLMs for Spatio-Temporal Reasoning in Time Series via Spatial-Aware Reinforcement Learning (2026.acl-long)

Copied to clipboard

Challenge: Existing models focus on predictive accuracy over reasoning, a gap exists . time series data are ubiquitous in real-world systems and exhibit complex spatio-temporal structures.
Approach: They propose a time series reasoning model that integrates time series, graph structure, and text for explicit reasoning.
Outcome: The proposed model achieves average accuracy gains between 17% and 135% at 0.004x the cost of proprietary models and generalizes robustly to real-world data.
Context-Aware Conversation Thread Detection in Multi-Party Chat (D19-1)

Copied to clipboard

Challenge: In multi-party chat, it is common for multiple conversations to occur concurrently . a new model that automatically disentangles conversation threads is proposed .
Approach: They propose a Context-Aware Thread Detection model that automatically disentangles conversation threads in chat logs.
Outcome: The proposed model outperforms state-of-the-art models on four real-world chat logs.
MCPEval: Automatic MCP-based Deep Evaluation for AI Agent Models (2025.emnlp-demos)

Copied to clipboard

Challenge: Existing evaluation frameworks suffer from limitations such as static task benchmarks, limited scope, and inadequate integration with practical applications.
Approach: They propose an open-source, Model Context Protocol-based evaluation framework specifically tailored for comprehensive and systematic assessment of LLM-powered agents.
Outcome: The proposed framework uncovers nuanced performance patterns and identify domain-specific strengths and weaknesses, providing valuable insights beyond traditional binary success metrics.
KG-Adapter: Enabling Knowledge Graph Integration in Large Language Models through Parameter-Efficient Fine-Tuning (2024.findings-acl)

Copied to clipboard

Challenge: Large language models (LLMs) are criticized for lack of expertise and knowledge conflict . KG-Adapter is a parameter-level KG integration method for decoder-only LLMs .
Approach: They propose a parameter-level KG integration method based on parameter-efficient fine-tuning . they use KG-Adapter to integrate knowledge graphs with LLMs and perform joint reasoning .
Outcome: The proposed method outperforms the current state-of-the-art method on four datasets for two different tasks.
CENTAUR: Bridging the Impossible Trinity of Privacy, Efficiency, and Performance in Privacy-Preserving Transformer Inference (2025.acl-long)

Copied to clipboard

Challenge: Existing privacy-preserving Transformer Inference frameworks suffer from high computational overhead and performance losses.
Approach: They propose a framework that integrates random permutations and SMPC to address the "impossible trinity" CENTAUR resists diverse data reconstruction attacks and boosts inference speed by 5.030.4 times .
Outcome: CENTAUR achieves an unprecedented balance between privacy, efficiency, and performance.
A Co-Matching Model for Multi-choice Reading Comprehension (P18-2)

Copied to clipboard

Challenge: Existing approaches to machine comprehension are based on pairwise sequence matching, but this approach is not suitable for multi-choice reading comprehension since questions and answers are often equally important.
Approach: They propose a co-matching approach that models whether a passage can match both a question and a candidate answer using a dataset from Chinese exams.
Outcome: The proposed approach achieves state-of-the-art on the RACE dataset from Chinese middle and high school English examinations.
Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning (2026.acl-long)

Copied to clipboard

Challenge: Recent work on Chain-of-Thought prompting imposes substantial computational overhead . lack of supervision obscures the analyzability of the latent reasoning chain.
Approach: They propose a framework to render latent reasoning chain into images, making latent rationale explicit and traceable.
Outcome: The proposed framework achieves 3-4 token compression and substantial inference acceleration compared to explicit CoT prompting.
Imposing Label-Relational Inductive Bias for Extremely Fine-Grained Entity Typing (N19-1)

Copied to clipboard

Challenge: Existing entity typing systems exploit type hierarchy provided by KB schema to model label correlations.
Approach: They propose a graph layer that encodes global label co-occurrence statistics and word-level similarities.
Outcome: The proposed model achieves a 15.3% relative F1 improvement on a large dataset with over 10,000 free-form types.
TWEETQA: A Social Media Focused Question Answering Dataset (P19-1)

Copied to clipboard

Challenge: Social media is becoming an important realtime information source, especially during natural disasters and emergencies.
Approach: They present a large-scale dataset for question answering over social media data . they gather tweets used by journalists and ask human annotators to write questions upon them .
Outcome: The proposed dataset shows that neural models that perform well on formal texts are limited in their performance . the proposed model is still lagging behind human performance with a large margin .
Pre-trained Semantic Interaction based Inductive Graph Neural Networks for Text Classification (2025.coling-main)

Copied to clipboard

Challenge: Existing methods for text classification have vanishing or exploding gradients when dealing with long sequences, making it difficult to handle long-distance dependencies.
Approach: They propose a graph neural network based on pre-trained semantic interaction called PaSIG . they construct a text-word heterogeneity graph and use context representation capability .
Outcome: The proposed model outperforms existing methods on five datasets and achieves state-of-the-art performance.
EvoMemKG: An Evolvable Memory Agent for Multi-hop Knowledge Graph Reasoning (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for integrating knowledge graphs with large language models lack continuous learning capabilities.
Approach: They propose an agent framework with a dynamic, evolvable memory mechanism specifically designed for KG reasoning.
Outcome: EvoMemKG achieves state-of-the-art performance without training or tools . it achieves improvements of up to 20% over baseline on multi-hop queries .
Diverse Few-Shot Text Classification with Multiple Metrics (N18-1)

Copied to clipboard

Challenge: Existing methods for few-shot learning are insufficient to capture task variations in natural language domains.
Approach: They propose an adaptive metric learning approach that automatically determines the best weighted combination from a set of metrics obtained from meta-training tasks for a newly seen few-shot task.
Outcome: The proposed method performs favorably against state-of-the-art few shot learning algorithms on real-world sentiment analysis and dialog intent classification datasets.

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