Papers by Haifeng Chen

36 papers
Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation (2021.emnlp-main)

Copied to clipboard

Challenge: Recent pre-trained language models have achieved remarkable zero-shot performance . we propose a self-learning framework that utilizes unlabeled data of target languages .
Approach: They propose a self-learning framework that utilizes unlabeled data of target languages to select silver labels for cross-lingual transfer tasks.
Outcome: The proposed framework outperforms baseline models on two cross-lingual tasks by 10 F1 on average and 2.5 accuracy on natural language inference (NLI).
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.
From Synthesis to Clinical Assistance: A Strategy-Aware Agent Framework for Autism Intervention based on Real Clinical Dataset (2026.acl-long)

Copied to clipboard

Challenge: Applied Behavior Analysis (ABA) is the gold standard for clinical intervention, but large language models struggle to adhere to its standardized procedures.
Approach: They propose a strategy-aware framework to unify high-fidelity intervention dialogue synthesis and clinical decision support.
Outcome: Experiments show that ASDAgent achieves nearly 80% strategic consistency with human experts.
Distributional Clarity: The Hidden Driver of RL-Friendliness in Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: RL-friendly models exhibit intra-class compactness and inter-class separation in probability assignments . under identical training, Qwen models achieve substantial gains, while others like Llama yield limited improvements.
Approach: They propose a method to quantify distributional clarity in probability space . they show distributional clearness is a trainable property underlying RL-Friendliness .
Outcome: The proposed model families achieve substantial gains under identical training, while others like Llama yield limited improvements.
Distantly-Supervised Joint Extraction with Noise-Robust Learning (2024.findings-acl)

Copied to clipboard

Challenge: Existing approaches to identifying entity pairs and relations with a single model are noisy . Existing methods only consider one source of noise or make decisions using external knowledge .
Approach: They propose a framework that aligns entity mentions with corresponding tags for joint extraction . they propose DENRL, which employs a lightweight transformer backbone for joint tagging .
Outcome: The proposed framework outperforms baseline models on two benchmark datasets with better interpretability.
Uncertainty-Aware Bootstrap Learning for Joint Extraction on Distantly-Supervised Data (2023.acl-short)

Copied to clipboard

Challenge: Existing approaches to extract entity pairs and their relations from labeled data are noisy and expensive.
Approach: They propose a bootstrap learning approach that is motivated by intuition that the higher the uncertainty of an instance, the more likely the model confidence is inconsistent with the ground truths.
Outcome: The proposed method outperforms baselines and related methods on two large datasets.
Uncertainty-Aware Test-Time Search for Optimization Problem Solving (2026.acl-long)

Copied to clipboard

Challenge: Language model hallucinations and limited availability of labeled datasets often result in misaligned formulations, code errors and feasibility failures.
Approach: They propose a Monte Carlo Tree Search framework that automates optimization problems from natural language descriptions with efficiency and reliability.
Outcome: The proposed framework achieves state-of-the-art solution accuracy and reduces token usage.
RLKGF: Reinforcement Learning from Knowledge Graph Feedback Without Human Annotations (2025.findings-acl)

Copied to clipboard

Challenge: Lack of human preference labels remains a significant bottleneck when applying RLHF to a downstream domain.
Approach: They propose a method that leverages human priors encoded in Knowledge Graphs (KGs) to derive RL rewards in the absence of manual annotations.
Outcome: Experiments on three public and one private medical dialogue datasets show that the proposed method outperforms the competitive RLAIF in improving LLM diagnostic accuracy.
Unsupervised Concept Representation Learning for Length-Varying Text Similarity (2021.naacl-main)

Copied to clipboard

Challenge: Existing document similarity approaches suffer from the information gap caused by context and vocabulary mismatches when comparing varying-length texts.
Approach: They propose an unsupervised concept representation learning approach to address this issue . they propose a concept-based document matching method to leverage recognition of local phrase features .
Outcome: The proposed method achieves a better F1 score than baseline models on real-world data sets.
DuReader-Retrieval: A Large-scale Chinese Benchmark for Passage Retrieval from Web Search Engine (2022.emnlp-main)

Copied to clipboard

Challenge: Existing datasets for non-English passage retrieval are lacking in quality and accuracy.
Approach: They present a large-scale Chinese dataset for passage retrieval . they reduce false negatives by manually annotating results pooled from multiple retrievers .
Outcome: The proposed dataset reduces false negatives in development and testing sets and removes similar training queries.
Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Existing cultural alignment approaches fail to align LLMs’ broad cultural values with the specific goals of downstream tasks and suffer from cross-cultural interference.
Approach: They propose a novel pipeline for task-specific cultural alignment that synthesizes task-aware cultural data in line with target task formats.
Outcome: Experiments across five national cultures and ten culture-sensitive tasks show consistent improvements over prompt-based and fine-tuning baselines.
Decoding Time Series with LLMs: A Multi-Agent Framework for Cross-Domain Annotation (2026.findings-eacl)

Copied to clipboard

Challenge: Time series data is ubiquitous across various domains, including manufacturing, finance, and healthcare.
Approach: They propose a multi-agent system to generate general and domain-specific annotations for time series data.
Outcome: The proposed system outperforms existing methods on synthetic and real-world datasets.
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.
Representation Interventions Enable Lifelong Knowledge Memory Control in LLMs (2026.acl-long)

Copied to clipboard

Challenge: Large language models (LLMs) produce outdated or inaccurate content. Updating their knowledge efficiently and accurately without costly retraining is a major challenge.
Approach: They propose a robust and scalable method that treats knowledge control as interventions within the model’s representation space.
Outcome: The proposed method achieves fine-grained control over complex, unstructured knowledge while maintaining general utility with frozen base weights.
Pruning as a Domain-specific LLM Extractor (2024.findings-naacl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have exhibited remarkable proficiency across a wide array of NLP tasks.
Approach: They propose a method for pruning large language models using general or task-specific weights to extract a compressed, task-agnostic LLM.
Outcome: The proposed method extracts a compressed, domain-specific, and task- agnostic LLM by identifying LLM weights that are pivotal for general capabilities, like linguistic capability and multi-task solving, and domain- specific knowledge.
InfuserKI: Enhancing Large Language Models with Knowledge Graphs via Infuser-Guided Knowledge Integration (2024.findings-emnlp)

Copied to clipboard

Challenge: Large Language Models have exceptional capabilities in open generation, yet they encounter difficulties with tasks that require intensive knowledge.
Approach: They propose a framework that integrates unknown knowledge into LLMs without overlap . they propose integrating domain-specific knowledge graphs into Llms to reduce knowledge forgetting .
Outcome: The proposed framework outperforms state-of-the-art baselines in integrating new knowledge into LLMs.
SgSum:Transforming Multi-document Summarization into Sub-graph Selection (2021.emnlp-main)

Copied to clipboard

Challenge: Existing extractive multi-document summarization methods score each sentence individually and extract salient sentences one by one.
Approach: They propose a novel framework for extractive multi-document summarization that selects a sub-graph as the summary instead of selecting salient sentences.
Outcome: The proposed framework improves on existing methods on multi-document datasets and human evaluations show it produces more coherent and informative summaries.
Position Really Matters: Towards a Holistic Approach for Prompt Tuning (2025.findings-naacl)

Copied to clipboard

Challenge: Prompt tuning is effective in extracting knowledge from foundation models, but its effectiveness is uncertain.
Approach: They propose a parametric prompt tuning strategy that dynamically determines different factors of prompts based on specific tasks or instances.
Outcome: The proposed approach improves performance across a wide range of tasks including NLP, vision recognition, and vision-language tasks.
Open-ended Commonsense Reasoning with Unrestricted Answer Candidates (2023.findings-emnlp)

Copied to clipboard

Challenge: Current approaches to commonsense reasoning are limited due to limited answer scope.
Approach: They propose to solve a commonsense question without a pre-defined answer scope . they leverage pre-trained language models to iteratively retrieve reasoning paths on the external knowledge base .
Outcome: The proposed method achieves better performance on two commonsense benchmark datasets.
Stop When Enough: Adaptive Early-Stopping for Chain-of-Thought Reasoning (2026.acl-long)

Copied to clipboard

Challenge: Chain-of-Thought reasoning has driven recent gains of large language models (LLMs) on reasoning-intensive tasks by externalizing intermediate steps.
Approach: They propose a training-free framework that adaptively determines when to stop reasoning to mitigate overthinking.
Outcome: The proposed framework reduces token usage by 20-55% while maintaining or improving accuracy compared to standard CoT prompting.
A Survey on Detection of LLMs-Generated Content (2024.findings-emnlp)

Copied to clipboard

Challenge: Recent advances in large language models have led to an increase in synthetic content generation . the ability to detect LLMs-generated content has become of paramount importance .
Approach: They propose to provide a detailed overview of existing detection strategies and benchmarks, scrutinizing their differences and advocating for more adaptable and robust models to enhance detection accuracy.
Outcome: The proposed model will be able to detect human-written content in real time.
DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router (2026.findings-eacl)

Copied to clipboard

Challenge: Existing RAG methods lack fine-grained control over query and source sides, resulting in noisy retrieval and shallow reasoning.
Approach: They propose an agentic RAG framework that integrates information sieving via LLM-as-a-knowledge-router.
Outcome: Experiments on multi-hop QA tasks across heterogeneous sources demonstrate improved reasoning depth, retrieval precision, and interpretability over conventional approaches.
Recommend for a Reason: Unlocking the Power of Unsupervised Aspect-Sentiment Co-Extraction (2021.findings-emnlp)

Copied to clipboard

Challenge: Existing review-based recommenders favor large and complex language encoders that can only learn latent and uninterpretable text representations.
Approach: They propose a tightly coupled two-stage approach to extract latent user sentiments and item properties from reviews and an Attention-Property-aware Rating Estimator (APRE).
Outcome: Extensive experiments on seven real-world Amazon review datasets show that the proposed approach extracts the latent user sentiments, item properties, and the complicated interactions between the two components.
Uncertainty Quantification for In-Context Learning of Large Language Models (2024.naacl-long)

Copied to clipboard

Challenge: Existing studies on in-context learning have focused on quantifying the uncertainty associated with the model's response, but they neglect the complexity of the LLM and the uniqueness of in-constitut learning.
Approach: They propose a method to quantify the uncertainty associated with in-context learning and propose corresponding estimation method to quantify both types of uncertainties.
Outcome: The proposed method offers an unsupervised way to understand the prediction of in-context learning in a plug-and-play fashion.
Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement (2026.findings-eacl)

Copied to clipboard

Challenge: Recent advances in large language models (LLMs) show potential for graph extraction, but often yield ill-formed structures or misinterpret logical constructs such as gateways.
Approach: They propose a framework that treats procedural graph extraction as a multi-round reasoning process with structural and logical refinement agents.
Outcome: The proposed framework achieves significant improvements in structural correctness and logical consistency over strong baselines.
Evaluating the Expressive Appropriateness of Speech in Rich Contexts (2026.acl-long)

Copied to clipboard

Challenge: Existing methods for evaluating expressive speech focus on word accuracy, naturalness, signal quality, or emotional intensity at the utterance level.
Approach: They propose a framework for Evaluating Expressive Appropriateness in speech that assesses whether a speech sample aligns with the underlying communicative intent implied by its discourse-level narrative context.
Outcome: The proposed framework outperforms existing speech evaluation and analysis systems on a human-annotated test set.
Towards Interpretable Clinical Diagnosis with Bayesian Network Ensembles Stacked on Entity-Aware CNNs (2020.acl-main)

Copied to clipboard

Challenge: a novel framework for text-based diagnosis of diseases requires appropriate balance between accuracy and interpretability.
Approach: They propose a framework that stacks Bayesian Network Ensembles on top of CNN to build an accurate yet interpretable diagnosis system.
Outcome: The proposed framework outperforms the previous automatic diagnosis methods in accuracy performance and the diagnosis explanation of the framework is reasonable.
Uncertainty Propagation on LLM Agent (2025.acl-long)

Copied to clipboard

Challenge: Existing methods for estimating uncertainty in large language models (LLMs) focus on final-step outputs, which fail to account for cumulative uncertainty over multi-step decision-making process and dynamic interactions between agents and their environments.
Approach: They propose a framework that propagates uncertainty through each step of an LLM-based agent’s reasoning process.
Outcome: Extensive experiments on benchmark datasets show that the proposed framework outperforms state-of-the-art methods by 20%.
Large Language Models Can Be Contextual Privacy Protection Learners (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated remarkable linguistic comprehension and generation capability, but when applied to specialized industries, they face challenges such as hallucination, insufficient domain knowledge, and failing to incorporate the latest domain knowledge.
Approach: They propose a paradigm for fine-tuning LLMs that effectively injects domain-specific knowledge while safeguarding inference-time data privacy.
Outcome: The proposed model protects private data while enhancing the model's knowledge.
DuQM: A Chinese Dataset of Linguistically Perturbed Natural Questions for Evaluating the Robustness of Question Matching Models (2022.emnlp-main)

Copied to clipboard

Challenge: a comprehensive evaluation of QM models should be conducted on natural texts, not on artificial adversarial examples . ral models are often not robust to adversarials, which means they predict unexpected outputs .
Approach: They use a Chinese dataset to evaluate the robustness of QM models . they show that the effect of artificial adversarial examples does not work on natural texts .
Outcome: The proposed model is more robust than other models on natural questions with 32 linguistic perturbations.
MixLLM: Dynamic Routing in Mixed Large Language Models (2025.naacl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) exhibit potential artificial generic intelligence, however, their usage is costly with high response latency.
Approach: They develop a dynamic contextual-bandit-based routing system for query-LLM assignment that leverages query tags to enhance query embeddings.
Outcome: The proposed model maximizes response quality and minimizes cost and latency.
Exploring Multi-Modal Data with Tool-Augmented LLM Agents for Precise Causal Discovery (2025.findings-acl)

Copied to clipboard

Challenge: Existing statistical causal discovery methods rely on observational data and often overlook the semantic cues inherent in cause-and-effect relationships.
Approach: They propose a multi-agent system powered by tool-augmented Large Language Models that can combine data from multiple modalities and integrate multi-modal data for knowledge-driven reasoning.
Outcome: The proposed system has two agents: a Data Augmentation agent that retrieves and processes modality-augmented data, and a Causal Constraint agent that integrates multi-modal data for knowledge-driven reasoning.
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.
Graph-Based Chain-of-Thought Pruning for Reducing Redundant Reflections in Reasoning LLMs (2026.findings-acl)

Copied to clipboard

Challenge: Extending CoT through RL can induce undesirable thinking patterns such as overthinking . prior work has focused on inefficient reflection, which manifests in two problematic patterns: Indiscriminate Reflection and Repetitive Reflectione .
Approach: They propose a graph-based approach to optimize CoT by pruning each linear CoT into a directed acyclic graph with explicit dependency edges.
Outcome: The proposed approach reduces the average reasoning tokens by 42% while maintaining or improving accuracy.
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.
LOKA: Conflict-Aware LLM Knowledge Update with Adaptive Knowledge Memory (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches that tackle unlearning and learning separately encounter *task conflicts* and *knowledge management issues* when applied to comprehensive knowledge updates.
Approach: They propose a conflict-aware framework for Large language mOdel Knowledge updAtes that integrates updated knowledge across multiple memory units during training and integrates it with original LLM.
Outcome: The proposed framework is based on theoretical analysis and empirical evidence and validates the proposed framework with empirical and theoretical evidence.

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