Papers by Zhengzhong Liu

12 papers
Efficient (Soft) Q-Learning for Text Generation with Limited Good Data (2022.findings-emnlp)

Copied to clipboard

Challenge: Maximum likelihood estimation (MLE) is the predominant method for training text generation models.
Approach: They propose a new RL formulation for text generation from the soft Q-learning perspective using path consistency learning to combine the best of on-/off-policy updates and learn effectively from sparse reward.
Outcome: The proposed approach outperforms MLE and previous RL methods in a wide range of tasks.
Decentralized Arena: Towards Democratic and Scalable Automatic Evaluation of Language Models (2026.acl-long)

Copied to clipboard

Challenge: closed-ended question-based benchmarks struggle with saturation as newer models emerge . crowd-sourced leaderboards rely on costly and slow human judges .
Approach: They propose a framework that leverages collective intelligence from all large language models to evaluate each other.
Outcome: a new framework enables a democratic, pairwise evaluation of all large language models . it achieves 97% correlation with human judgements, while significantly reducing the cost.
Texar: A Modularized, Versatile, and Extensible Toolkit for Text Generation (P19-3)

Copied to clipboard

Challenge: Texar is an open-source text generation toolkit that supports a broad set of text generation tasks.
Approach: They introduce Texar, an open-source text generation toolkit that supports text generation tasks.
Outcome: Texar supports machine translation, summarization, dialog, content manipulation, and more.
Does RLVR Extend Reasoning Boundaries? Investigating Capability Expansion in Vision-Language Models (2026.acl-long)

Copied to clipboard

Challenge: Recent studies suggest that RLVR amplifies behaviors inherent to the pre-training distribution rather than inducing new capabilities.
Approach: They propose a framework for RLVR that extends the spatial reasoning boundary . they use a mapping framework where the difficulty is precisely regulated by path length and number of turns .
Outcome: The proposed framework extends the spatial reasoning boundary on two real-world navigation benchmarks.
Token Level Routing Inference System for Edge Devices (2025.acl-demo)

Copied to clipboard

Challenge: Large language models (LLMs) have been gaining in performance but deployment in edge devices faces significant hurdles due to their high computational complexity.
Approach: They propose a collaborative decoding system that allows small models to perform on-device inference while selectively consulting a cloud-based large model for critical token generation.
Outcome: The proposed system achieves 60% performance gain on CommonsenseQA using a 0.5B model on an M1 MacBook, with under 7% of tokens generation uploaded to the large model in the cloud.
Linear Steerability in Language Models: When It Emerges and How It Evolves (2025.findings-emnlp)

Copied to clipboard

Challenge: a new framework for steering language models reveals how concepts become linearly separable as training progresses .
Approach: They propose a framework to analyze steerability in language models by using hidden state and representation analysis.
Outcome: The proposed framework reveals how steerability evolves over training . concepts become linearly separable as training progresses, the framework shows .
Graph Based Decoding for Event Sequencing and Coreference Resolution (C18-1)

Copied to clipboard

Challenge: In this paper, we study two types of relation between events in text documents.
Approach: They propose a graph-based decoding algorithm that is applicable to both tasks . they propose ES and EH to solve the event coreference problem .
Outcome: The proposed decoding algorithm beats a strong temporal-based, oracle-informed baseline.
A Data-Centric Framework for Composable NLP Workflows (2020.emnlp-demos)

Copied to clipboard

Challenge: Empirical natural language processing (NLP) systems involve interoperation among multiple components . a wealth of NLP toolkits exist ( 4), such as spaCy, DKPro, CoreNLP.
Approach: They propose a unified open-source framework that supports fast development of NLP workflows . framework includes processors for NLP tasks, visualization, and annotation .
Outcome: The framework offers processors for NLP tasks, visualization, and annotation, and is extensible . it is delivered through two modularized yet integratable open-source projects, Forte and Stave .
A Two-Step Approach for Implicit Event Argument Detection (2020.acl-main)

Copied to clipboard

Challenge: et al., 2015) only consider local arguments in the same sentence of the event trigger.
Approach: They propose to decompose the implicit event argument detection task into two sub-problems . they propose to use argument head-word detection and head-to-span expansion to reduce the number of candidates.
Outcome: The proposed model achieves better performance than a strong sequence labeling baseline.
Automatic Event Salience Identification (D18-1)

Copied to clipboard

Challenge: Existing models for analyzing salience of discourse units are inadequate . authors propose two saliency detection models based on discourse relations .
Approach: They propose two salience detection models based on discourse relations that capture complex interactions between discourse units.
Outcome: The proposed models outperform the strong frequency baseline and improve the feature based model by a large margin.
Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation (2021.emnlp-main)

Copied to clipboard

Challenge: Natural language generation (NLG) tasks have complex nature and require manual evaluation.
Approach: They propose a unifying perspective based on the nature of information change in NLG tasks . they propose 'information alignment' metrics that can be used to evaluate different aspects of NLG .
Outcome: The proposed metrics achieve stronger or comparable correlations with human judgement compared to state-of-the-art metrics in diverse tasks.
ASDOT: Any-Shot Data-to-Text Generation with Pretrained Language Models (2022.findings-emnlp)

Copied to clipboard

Challenge: Existing approaches to data-to-text generation require limited training examples . a data-based approach is based on a set of pre-trained language models with optional finetuning.
Approach: They propose a data-to-text generation task that makes use of any given (or no) examples.
Outcome: The proposed approach improves on baselines on a dataset with zero/few/full-shot settings.

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