Papers by Zhenyu Wu

17 papers
Robust Machine Reading Comprehension by Learning Soft labels (2020.coling-main)

Copied to clipboard

Challenge: Neural models have achieved great success on the task of machine reading comprehension, which are typically trained on hard labels.
Approach: They propose a robust training method for machine reading comprehension models to address label sparseness problem by using three strategies to train models on soft labels.
Outcome: The proposed method improves the baseline model performance and achieves state-of-the-art performance on NewsQA and QUOREF.
Uncertainty-Aware Routing for Principled Alignment with MoE Dynamics (2026.acl-long)

Copied to clipboard

Challenge: Mixture-of-Experts (MoE) is a cornerstone for scaling LLMs, yet its training dynamics remain poorly understood, often leading to sub-optimal specialization.
Approach: They propose to use Helmholtz Free Energy and Router Entropy to study the MoE lifecycle and identify a universal Three-Stage Phase Transition .
Outcome: The proposed model reduces perplexity and improves expert distinctiveness, offering a principled path toward thermodynamically aligned computation.
CodeTaxo: Enhancing Taxonomy Expansion with Limited Examples via Code Language Prompts (2025.findings-acl)

Copied to clipboard

Challenge: Existing taxonomies are mainly constructed by experts or through crowd-sourcing, making the process time-consuming, labor-intensive, and restricted in coverage.
Approach: They propose a method that leverages large language models to capture taxonomic structure . existing taxonomies are mainly constructed by experts or through crowd-sourcing .
Outcome: Experiments on five real-world domains show that CodeTaxo outperforms state-of-the-art methods.
LEMON: Reviving Stronger and Smaller LMs from Larger LMs with Linear Parameter Fusion (2024.acl-long)

Copied to clipboard

Challenge: Existing methods to train a stronger and smaller model with the help of large models are limited by the model size and performance.
Approach: They propose to learn competent initial points for smaller models by fusing parameters from larger models and introduce controllable receptive fields to model prior parameter characteristics.
Outcome: The proposed method outperforms baselines in terms of effectiveness and efficiency.
Enhancing Mathematical Reasoning in LLMs by Stepwise Correction (2025.acl-long)

Copied to clipboard

Challenge: Existing Best-of-N decoding methods often lead to incorrect solutions . a novel method is proposed to help large language models identify and revise incorrect steps in their generated reasoning paths.
Approach: They propose a method that helps large language models identify and revise incorrect steps in their generated reasoning paths.
Outcome: The proposed method outperforms the state-of-the-art Best-ofN decoding method by +2.4 and reduces token consumption by 77.8%.
Inner Thinking Transformer: Leveraging Dynamic Depth Scaling to Foster Adaptive Internal Thinking (2025.acl-long)

Copied to clipboard

Challenge: Large language models face inherent performance bottlenecks under parameter constraints . challenging tokens induce abrupt gradient spikes across layers, exposing stress points .
Approach: They propose an inner thinking transformer that reimagines layer computations as implicit thinking steps.
Outcome: Empirical results show that ITT outperforms Transformer/Loop variants in 11 benchmarks.
NACL: A General and Effective KV Cache Eviction Framework for LLM at Inference Time (2024.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) with extended context windows are expensive and infeasible on fixed memory hardware due to the surprisingly large memory consumption of KV Cache.
Approach: They propose a general framework for long-context KV cache eviction that achieves more optimal and efficient evict in a single operation during the encoding phase.
Outcome: The proposed framework improves performance on short- and long-text tasks by 80% and 76% respectively, reducing KV Cache by up to 5 with over 95% performance maintenance.
OS-Symphony: A Holistic Framework for Robust and Generalist Computer-Using Agents (2026.acl-long)

Copied to clipboard

Challenge: Vision-Language Models (VLMs) lack visual-aware tutorial retrieval and historical visual context curation and pruning.
Approach: They propose a framework that integrates an orchestrator and a Reflection-Memory Agent for robust automation.
Outcome: Experimental results show that OS-Symphony delivers substantial performance gains across model scales.
OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse Task Synthesis (2025.acl-long)

Copied to clipboard

Challenge: Graphical User Interface (GUI) agents powered by Vision-Language Models (VLMs) have demonstrated human-like computer control capability.
Approach: They propose a GUI data synthesis pipeline that reverse engineers GUI trajectory construction process by executing pre-defined tasks.
Outcome: The proposed GUI data synthesis pipeline overcomes the bottlenecks of previous methods that rely on pre-defined tasks and limited data diversity.
HFT: Half Fine-Tuning for Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) with one or more fine-tuning phases can unlock various capabilities, but can be catastrophic forgetting during sequential training.
Approach: They propose a method to regularly reset partial parameters to mitigate forgetting issues by using half fine-tuning instead of full fine-uning.
Outcome: The proposed approach reduces the risk of catastrophic forgetting during training and the parametric knowledge lost during training may be overwhelmed by incoming training data.
Large Language Models Can Self-Correct with Key Condition Verification (2024.emnlp-main)

Copied to clipboard

Challenge: Existing methods to correct reasoning without external feedback have not been used in large language models.
Approach: They propose an iterative verify-then-correct framework to progressively identify and correct (probably) false responses, named ProCo.
Outcome: The proposed method improves the accuracy of LLMs on three reasoning tasks.
Sparse Growing Transformer: Training-Time Sparse Depth Allocation via Progressive Attention Looping (2026.findings-acl)

Copied to clipboard

Challenge: Existing approaches to increasing effective depth of LLMs rely on parameter reuse, extending computation through recursive execution.
Approach: They propose a training-time sparse depth allocation framework that progressively increases depth for a small subset of parameters as training evolves.
Outcome: The proposed model outperforms existing approaches to increasing the effective depth of language models while reducing training FLOPs overhead from approximately 16–20% to only 1–3% relative to a standard Transformer backbone.
BeamLoRA: Beam-Constraint Low-Rank Adaptation (2025.acl-long)

Copied to clipboard

Challenge: Low-Rank Adaptation (LoRA) is one of the most efficient parameter-efficient fine-tuning methods.
Approach: They propose to conceptualize each LoRA module as a beam where each rank corresponds to a potential sub-solution.
Outcome: The proposed method improves performance on three base models and 12 datasets.
Instructing Large Language Models to Identify and Ignore Irrelevant Conditions (2024.naacl-long)

Copied to clipboard

Challenge: Existing CoT prompting methods elicited multi-step reasoning abilities of large language models (LLMs) but they were seriously confused by the irrelevant conditions, resulting in low accuracy.
Approach: They propose a method that instructs large language models to identify and ignore irrelevant conditions and prompts them to verify the irrelevant conditions.
Outcome: The proposed approach outperforms existing methods on MWPs with GPT-3.5-Turbo and I3C-Select.
Upcycling Instruction Tuning from Dense to Mixture-of-Experts via Parameter Merging (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for tuning large language models from dense to MoE face significant data requirements and require large-scale post-training.
Approach: They propose an upcycling instruction tuning approach for tuning a dense pre-trained model into a MoE instruction model using genetic algorithm and parameter merging.
Outcome: The proposed approach improves the performance of large language models with a small amount of seed data and improves their scaling.
OpenICL: An Open-Source Framework for In-context Learning (2023.acl-demo)

Copied to clipboard

Challenge: In-context Learning (ICL) is a new paradigm for large language model evaluation.
Approach: They propose an open-source toolkit for ICL and LLM evaluation.
Outcome: The proposed framework is highly flexible and flexible and can be easily combined with other tools to suit users' needs.
ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for visually rich document understanding lack layout-centered knowledge . experimental results show that ERNIE-Layout improves layout awareness .
Approach: They propose a document pre-training solution with layout knowledge enhancement in the whole workflow to learn better representations that combine the features from text, layout, and image.
Outcome: The proposed model outperforms existing models on key downstream tasks.

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