Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks192 from public data
- A Survey on In-context Learning1,205
This paper presents a formal definition of ICL and clarify its correlation to related studies, and organizes and discusses advanced techniques, including training strategies, prompt designing strategies, and related analysis.
- Knowledge Neurons in Pretrained Transformers785
This paper examines the fill-in-the-blank cloze task for BERT and proposes a knowledge attribution method to identify the neurons that express the fact, finding that the activation of such knowledge neurons is positively correlated to the expression of their corresponding facts.
- Gated Self-Matching Networks for Reading Comprehension and Question Answering673
The gated self-matching networks for reading comprehension style question answering, which aims to answer questions from a given passage, are presented and holds the first place on the SQuAD leaderboard for both single and ensemble model.
- Double Graph Based Reasoning for Document-level Relation Extraction290
This paper proposes Graph Aggregation-and-Inference Network (GAIN) featuring double graphs, based on which GAIN first constructs a heterogeneous mention-level graph (hMG) to model complex interaction among different mentions across the document and proposes a novel path reasoning mechanism to infer relations between entities.
- Table-to-Text Generation by Structure-Aware Seq2seq Learning286
The attention visualizations and case studies show that the novel structure-aware seq2seq architecture which consists of field-gating encoder and description generator with dual attention is capable of generating coherent and informative descriptions based on the comprehensive understanding of both the content and the structure of a table.
- Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models263
A comprehensive and challenging benchmark specifically designed to assess LLMs' mathematical reasoning at the Olympiad level, which shows that even the most advanced models, OpenAI o1-mini and OpenAI o1-preview, struggle with highly challenging Olympiad-level problems, highlighting significant challenges in Olympiad-level mathematical reasoning.
- CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark262
The first Chinese Biomedical Language Understanding Evaluation Evaluation (CBLUE) benchmark is presented: a collection of natural language understanding tasks including named entity recognition, information extraction, clinical diagnosis normalization, single-sentence/sentence-pair classification, and an associated online platform for model evaluation, comparison, and analysis.
- Jointly Extracting Event Triggers and Arguments by Dependency-Bridge RNN and Tensor-Based Argument Interaction261
A novel dependency bridge recurrent neural network (dbRNN) is proposed that simultaneously applying tree structure and sequence structure in RNN brings much better performance than only uses sequential RNN.
- Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning235
A straightforward yet effective fine-tuning technique, Child-Tuning, which updates a subset of parameters (called child network) of large pretrained models via strategically masking out the gradients of the non-child network during the backward process is proposed.
- MMICL: Empowering Vision-language Model with Multi-Modal In-Context Learning224
This paper introduces vision-language Model with Multi-Modal In-Context Learning (MMICL), a new approach to allow the VLM to deal with multi-modal inputs efficiently and constructs the Multi-modal In-Context Learning (MIC) dataset, designed to enhance the VLM's ability to understand complex multi-modal prompts.
- 190
- Towards Time-Aware Knowledge Graph Completion184
A novel time-aware knowledge graph completion model that is able to predict links in a KG using both the existing facts and the temporal information of the facts and achieves the state-of-the-art on temporal facts consistently.
- Max-Margin Tensor Neural Network for Chinese Word Segmentation169
A novel neural network model for Chinese word segmentation called Max-Margin Tensor Neural Network (MMTNN), which has the ability to model complicated interactions between tags and context characters and a new tensor factorization approach is proposed to speed up the model and avoid overfitting.
- Graph-based Dependency Parsing with Bidirectional LSTM155
A neural network model for graph-based dependency parsing which utilizes Bidirectional LSTM (BLSTM) to capture richer contextual information instead of using high-order factorization, and enable the model to use much fewer features than previous work.
- 150
- A Soft-label Method for Noise-tolerant Distantly Supervised Relation Extraction150
This work introduces an entity-pair level denoise method which exploits semantic information from correctly labeled entity pairs to correct wrong labels dynamically during training, and proposes a joint score function which combines the relational scores based on the entity- Pair representation and the confidence of the hard label to obtain a new label.
- Exploiting collaborative filtering techniques for automatic assessment of student free-text responses144
The experimental results show that the feature-based matrix factorization model outperforms the baseline models and the model with a re-ranking phase can achieve a better and competitive performance -- 63.6% overall accuracy on the Beetle dataset.
- Encoding Temporal Information for Time-Aware Link Prediction136
A novel time-aware KB embedding approach taking advantage of the happening time of facts is proposed, which uses temporal order constraints to model transformation between time-sensitive relations and enforce the embeddings to be temporally consistent and more accurate.
- Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker134
GIT introduces a Tracker module to track the extracted events and hence capture the interdependency among the events in a document to solve the two challenges of document-level event extraction.
- Order-Planning Neural Text Generation From Structured Data117
This paper proposes an order-planning text generation model, where order information is explicitly captured by link-based attention and a self-adaptive gate combines the link- based attention with traditional content-based Attention.
- Incorporating Glosses into Neural Word Sense Disambiguation107
GAS models the semantic relationship between the context and the gloss in an improved memory network framework, which breaks the barriers of the previous supervised methods and knowledge-based methods.
- SIRE: Separate Intra- and Inter-sentential Reasoning for Document-level Relation Extraction105
This paper designs a new and straightforward form of logical reasoning module that can cover more logical reasoning chains and outperforms the previous state-of-the-art methods on document-level relation extraction.
- Discourse Parsing with Attention-based Hierarchical Neural Networks102
Experimental results show that the proposed attention-based hierarchical neural network model for discourse parsing obtains comparable performance to the contemporary state-of-the-art systems with little manual feature engineering.
- 81
- An Enhanced Span-based Decomposition Method for Few-Shot Sequence Labeling73
SD decomposes the span matching problem into a series of span-level procedures, mainly including enhanced span representation, class prototype aggregation and span conflicts resolution, which achieves the new state-of-the-art results on two popular FSSL benchmarks, FewNERD and SNIPS.
- Reading and Thinking: Re-read LSTM Unit for Textual Entailment Recognition68
A new L STM unit called re-read LSTM (rLSTM) is designed to implement this “thinking” process, which means to read the premise again and again while reading the hypothesis to get a better understanding of the premise.
- Leveraging Gloss Knowledge in Neural Word Sense Disambiguation by Hierarchical Co-Attention66
This paper introduces a co-attention mechanism to generate co-dependent representations for the context and gloss, and extends the attention mechanism in a hierarchical fashion in order to capture both word-level and sentence-level information.
- An Effective Neural Network Model for Graph-based Dependency Parsing62
This paper proposes a general and effective Neural Network model for graph-based dependency parsing that can automatically learn high-order feature combinations using only atomic features by exploiting a novel activation function tanhcube.
- Towards End-to-End Embodied Decision Making via Multi-modal Large Language Model: Explorations with GPT4-Vision and Beyond59
This study introduces a new benchmark called PCA-EVAL, which evaluates embodied decision-making from the perspectives of Perception, Cognition, and Action and proposes HOLMES, a multi-agent cooperation framework that allows LLMs to leverage MLLMs and APIs to gather multimodal information for informed decision- making.
- 58
- Learning to Control the Fine-grained Sentiment for Story Ending Generation54
To the best of the knowledge, this is the first endeavor to control the fine-grained sentiment for story ending generation without manually annotating sentiment labels.
- 52
- 50
- Towards Fine-grained Text Sentiment Transfer47
This paper proposes a novel Seq2SentiSeq model, where the numeric sentiment intensity value is incorporated into the decoder via a Gaussian kernel layer to finely control the sentiment intensity of the output.
- RBPB: Regularization-Based Pattern Balancing Method for Event Extraction46
Inspired by the progress in representation learning, trigger embedding, sentence-level embedding and pattern features together as their features for trigger classification are used so that the effect of patterns and other useful features can be balanced.
- 45
- Hierarchical Encoder with Auxiliary Supervision for Neural Table-to-Text Generation: Learning Better Representation for Tables45
A two-level hierarchical encoder with coarse-to-fine attention to handle the attribute-value structure of the tables, and proposes 3 joint tasks apart from the prime encoder-decoder learning, namely auxiliary sequence labeling task, text autoencoder and multi-labeling classification, as the auxiliary supervisions for the table encoder.
- 43
- Chinese Semantic Role Labeling with Bidirectional Recurrent Neural Networks41
Bidirectional recurrent neural network with long-short-term memory (LSTM) with RNN to capture bidirectional and long-range dependencies in a sentence with minimal feature engineering is introduced.
- Next Token Prediction Towards Multimodal Intelligence: A Comprehensive Survey38
This survey introduces a comprehensive taxonomy that unifies both understanding and generation within multimodal learning through the lens of NTP, and aims to aid researchers in their exploration of multimodal intelligence.
- Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View38
Two methods for faithful generation are proposed: augmented training by incorporating the auxiliary entity information, including both an augmented plan-based model and an unsupervised model and 2) training instance selection based on faithfulness ranking which improve generation fidelity in both full dataset setting and few shot setting.
- 37
- BabyVision: Visual Reasoning Beyond Language35
This work introduces BabyVision, a benchmark designed to assess core visual abilities independent of linguistic knowledge for Multimodal LLMs, and explores solving visual reasoning with generation models by proposing BabyVision-Gen and automatic evaluation toolkit.
- Pun-GAN: Generative Adversarial Network for Pun Generation35
Experiments show that the proposed Pun-GAN can generate sentences that are more ambiguous and diverse in both automatic and human evaluation.
- ERSOM: A Structural Ontology Matching Approach Using Automatically Learned Entity Representation35
An ontology matching approach, named ERSOM, which mainly includes an unsupervised representation learning method based on the deep neural networks to learn the general representation of the entities and an iterative similarity propagation method that takes advantage of more abundant structure information of the ontology to discover more mappings.
- From Dense to Sparse: Contrastive Pruning for Better Pre-trained Language Model Compression34
ContrAstive Pruning (CAP) is proposed, designed as a general framework, compatible with both structured and unstructured pruning, and enables the pruned model to learn from the pre-trained model for task-agnostic knowledge, and fine-tuned model fortask-specific knowledge.
- Human-in-the-Loop through Chain-of-Thought33
A Cost-utility Analysis Model for Human-in-the-Loop systems (CAMLOP) based on classical economics theory to analyze, quantify and balance the utility and the corresponding cost.
- Joint Learning Templates and Slots for Event Schema Induction31
A joint entity-driven model to learn template and slots simultaneously based on the constraints of templates and slots in the same sentence is proposed and the entities' semantic information is also considered for the inner connectivity of the entities.
- A Multi-View Fusion Neural Network for Answer Selection29
A Multi-View Fusion Neural Network, where each attention component generates a ``view'' of the QA pair and a fusion RNN integrates the generated views to form a more holistic representation.
- Large-Scale Simple Question Generation by Template-Based Seq2seq Learning28
A 28M Chinese Q&A corpora based on the Chinese knowledge base provided by NLPCC2017 KBQA challenge is presented and a novel neural network architecture which combines template-based method and seq2seq learning to generate highly fluent and diverse questions is proposed.
- An Empirical Study on Model-agnostic Debiasing Strategies for Robust Natural Language Inference26
This paper benchmarks prevailing neural NLI models including pretrained ones on various adversarial datasets and tries to combat distinct known biases by modifying a mixture of experts (MoE) ensemble method and shows that it’s nontrivial to mitigate multiple NLI biases at the same time, and that model-level ensemble method outperforms MoE ensemble method.
- Chinese Computational Linguistics and Natural Language Processing Based on Naturally Annotated Big Data26
A radical enhanced Chinese word embedding model that uses conversion and radical escaping mechanisms to extract the intrinsic information in Chinese corpus and has effectively improved the accuracy of word vector expression and the direct relevance of similar words.
- Syntax Aware LSTM model for Semantic Role Labeling26
Syntax Aware Long Short Time Memory is proposed, which can model the whole tree structure of dependency relation in an architecture engineering way and brings improvement to bi-LSTM model and also brings slight improvement when added to the state-of theart model.
- Bring you to the past: Automatic Generation of Topically Relevant Event Chronicles25
Experimental results demonstrate the first novel approach to automatically generate a topically relevant event chronicle during a certain period given a reference chronicles during another period is promising to tackle this new problem.
- Multi-label Text Categorization with Joint Learning Predictions-as-Features Method25
A novel joint learning algorithm is proposed that allows the feedbacks to be propagated from the classifiers for latter labels to the classifier for the current label, and the predictions-as-features models trained by the algorithm outperform the original models.
- Towards Comprehensive Description Generation from Factual Attribute-value Tables24
This work proposes force attention (FA) method to encourage the generator to pay more attention to the uncovered attributes to avoid potential key attributes missing, and proposes reinforcement learning for information richness to generate more informative as well as more loyal descriptions for tables.
- GATEAU: Selecting Influential Samples for Long Context Alignment23
GATEAU is proposed, a novel framework to address the unique challenge of long context alignment by identifying the influential samples enriched with long-range dependency relations and the model trained on these selected samples exhibits better instruction-following and long-context understanding capabilities.
- Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control is Easier than You Think23
Preliminary ex-periments show that large multimodal models (LMMs) offer an effective shared representation space, where image and text can be well-aligned to serve as a condition for external diffusion models, and propose Dream Engine, an efficient and unified frame-work designed for arbitrary text-image interleaved control in image generation models.
- G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning23
Systematic analysis reveals an intriguing finding: perception and reasoning abilities mutually bootstrap each other throughout the RL training process, allowing VLMs to outperform leading proprietary models like Claude-3.7-Sonnet-Thinking.
- Improving Chinese semantic role classification with hierarchical feature selection strategy23
A semantic role classifier based on a hierarchical feature selection strategy that outperforms the strong baseline significantly and is also the state-of-the-art on Chinese SRC is built.
- Event Detection with Burst Information Networks22
Two fast centroid-aware event detection models based on a novel text stream representation – Burst Information Networks (BINets) and BINet-based approaches are proposed for addressing the challenge of retrospective event detection.
- A novel topic model for automatic term extraction22
This paper proposes to compute termhood based on semantic representation of words based on i-SWB, a novel topic model developed to map the domain corpus into a latent semantic space, which is composed of some general topics, a background topic and a documents-specific topic.
- A Double-Graph Based Framework for Frame Semantic Parsing20
This paper proposes a Knowledge-guided Incremental semantic parser with Double-graph (KID), a heterogeneous graph containing both frames and FEs (Frame Elements) built on the frame knowledge so that it can derive knowledge-enhanced representations for frames andFEs.
- Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions20
This work proposes a simple and effective method that inductively represents OOKG entities by their optimal estimation under translational assumptions and shows that it outperforms the previous methods by a large margin with higher efficiency.
- Improved Dependency Parsing using Implicit Word Connections Learned from Unlabeled Data20
Experiments show that these implicit word connections do improve the parsing model and by combining with a pre-trained language model, the model gets state-of-the-art performance on the English PTB dataset.
- Affinity-Preserving Random Walk for Multi-Document Summarization20
This paper introduces affinity-preserving random walk to the summarization task, which preserves the affinity relations of sentences by an absorbing random walk model to enforce the diversity constraint of summarization in the random walk process.
- A Joint Model for Unsupervised Chinese Word Segmentation20
Analysis indicates that compared with nVBE and HDP, the joint model has a stronger ability to solve both combinational and overlapping ambiguities in Chinese word segmentation.
- Extraction of translation unit from Chinese-English parallel corpora20
An iterative algorithm based on degree of word association is proposed to identify the multiword units for Chinese and English and the Chinese-English Translation Equivalent Pairs are extracted from the parallel corpus.
- Update Summarization Using a Multi-level Hierarchical Dirichlet Process Model19
This paper borrows the idea of evolutionary clustering and proposes a three-level HDP model named h-uHDP, which reveals the diversity and commonality between aspects discovered from two different epochs (i.e. epoch history and epoch update).
- Hierarchical Curriculum Learning for AMR Parsing17
A Hierarchical Curriculum Learning (HCL) framework with Structure-level (SC) and Instance-level Curricula (IC) that reduces the difficulty of learning complex structures, thus the flat model can better adapt to the AMR hierarchy.
- A Roadmap for Big Model17
This paper discusses not only the BM technologies themselves but also the prerequisites for BM training and applications with BMs, dividing the BM review into four parts: Resource, Models, Key Technologies and Application.
- Mining Clues from Incomplete Utterance: A Query-enhanced Network for Incomplete Utterance Rewriting16
This work proposes a QUEry-Enhanced Network (QUEEN), which adopts a fast and effective edit operation scoring network to model the relation between two tokens and achieves state-of-the-art performance on several public datasets.
- SCL-RAI: Span-based Contrastive Learning with Retrieval Augmented Inference for Unlabeled Entity Problem in NER16
This paper decreases the distance of span representations with the same label while increasing it for different ones via span-based contrastive learning, which relieves the ambiguity among entities and improves the robustness of the model over unlabeled entities.
- Behind the Scenes16
To cope with the context-bypassing problem in FSEC models, adversarial training and trigger reconstruction techniques are introduced and experiments show these techniques help not only improve the performance, but also enhance the generalization ability of models.
- 15
- A Spectral Method for Unsupervised Multi-Document Summarization15
A spectral-based hypothesis is proposed, which states that the goodness of summary candidate is closely linked to its so-called spectral impact, and an accelerated greedy solution based on a surrogate of spectral impact is proposed.
- Chinese word segmentation at Peking University15
There is no need for an absolute definition of word boundary for all segmenters, and that different results of segmentation shall be acceptable if they can help to reach a correct syntactic analysis in the end.
- Step-wise Rubric Rewards for LLM Reasoning14
Step-wise Rubrics as Rewards is introduced, an RLVR framework that uses an LLM judge to attribute each rubric item to a specific reasoning step, normalizes per-step rubric scores across rollouts so only steps whose quality varies produce a learning signal, and combines the per-step reward with the outcome reward through a decoupled advantage estimator that keeps the outcome baseline stable.
- DiffCap: Exploring Continuous Diffusion on Image Captioning14
This work proposes a novel method DiffCap to apply continuous diffusions on image captioning to fuse multi-modal features in diffusion language generation and believes it will inspire more researches on multimodal language generation tasks for its simplicity and decoding flexibility.
- Premise-based Multimodal Reasoning: Conditional Inference on Joint Textual and Visual Clues14
Inspired by the designs of both visual commonsense reasoning and natural language inference tasks, a new task is proposed termed “Premise-based Multi-modal Reasoning” (PMR) where a textual premise is the background presumption on each source image.
- News Stream Summarization using Burst Information Networks14
Extensive experiments show that the BINet-based approaches are not only efficient and can be used in a real-time online summarization setting, but also can generate high-quality summaries, outperforming the state-of-the-art approach.
- A Maximum Entropy Approach to Chinese Spelling Check14
Experiment shows that by importing a large raw corpus, maximum entropy can build a well-trained model to detect spelling errors in Chinese documents.
- Chinese-English Parallel Corpus Construction and its Application14
The collection, annotation and mark-up of the parallel Chinese-English texts and the workflow that is used to construct the corpus are presented and a parallel concordance tool developed by us is examined in detail.
- Decompose, Fuse and Generate: A Formation-Informed Method for Chinese Definition Generation13
This paper builds a formation-informed dataset, and proposes a model DeFT, which Decomposes words into formation features, dynamically Fuses different features through a gating mechanism, and generaTes word definitions.
- Joint Multi-Decoder Framework with Hierarchical Pointer Network for Frame Semantic Parsing13
A novel architecture based on multi-decoder strategy to handle these subtasks together and a hierarchical pointer network for argument identification which reduces the computational complexity.
- Generating Math Word Problems from Equations with Topic Consistency Maintaining and Commonsense Enforcement12
This paper presents a novel equation-to-problem text generation model, which outperforms baseline and previous models in both accuracy and richness of generated problem text.
- 12
- LongR: Unleashing Long-Context Reasoning via Reinforcement Learning with Dense Utility Rewards11
This work proposes LongR, a unified framework that enhances long-context performance by integrating a dynamic ``Think-and-Read''mechanism---which interleaves reasoning with document consultation---with a contextual density reward based on relative information gain to quantify the utility of the relevant documents, without evidence annotations.
- Coarse-to-Fine Entity Representations for Document-Level Relation Extraction10
This work proposes the coarse-to-fine CFER model, a graph-based model that adopts a coarse- to-fine strategy involving two phases that combines the entity representations from both two levels into more comprehensive representations for relation extraction.
- Event-Based Time Label Propagation for Automatic Dating of News Articles10
The experiments show that the event-based propagation model can predict document timestamps in high accuracy and the model combined with a MaxEnt classifier outperforms the state-ofthe-art method for this task especially when the size of the training set is small.
- Enhancing Domain Portability of Chinese Segmentation Model Using Chi-Square Statistics and Bootstrapping10
This paper proposes to use X2 statistics when training an SVM-HMM based segmentation model to improve its ability to recall OOV words and then use bootstrapping strategies to maintain its able to recall IV words.
- A Spark of Vision-Language Intelligence: 2-Dimensional Autoregressive Transformer for Efficient Finegrained Image Generation9
This work tackles the information loss bottleneck of vector-quantization (VQ) autoregressive image generation by introducing a novel model architecture called the 2-Dimensional Autoregression (DnD) Transformer, an end-to-end model that can generate higher quality images with the same backbone model size and sequence length.
- 9
- 9
- DISK: Domain-constrained Instance Sketch for Math Word Problem Generation9
A neural model for generating MWP text from math equations that improves the model’s comprehension of real-world scenarios and derive a domain-constrained instance sketch to guide the generation.
- 9
- Recognizing Textual Entailment Using Probabilistic Inference9
This paper proposes to use the predicate-argument structures to represent the discourse commitments extracted from text and borrows the distant supervision technique to mine the implicit facts from the text.
- S4-Tuning: A Simple Cross-lingual Sub-network Tuning Method-Tuning: A Simple Cross-lingual Sub-network Tuning Method8
S^4-Tuning, a Simple Cross-lingual Sub-network Tuning method, which first detects the most essential sub-network for each target language, and only updates it during fine-tuning, to address two problems elegantly: language interference and Parameter Overload.
- Discriminatively-Tuned Generative Classifiers for Robust Natural Language Inference8
This paper proposes GenNLI, a generative classifier for NLI tasks, and empirically characterize its performance by comparing it to five baselines, including discriminative models and large-scale pretrained language representation models like BERT.
- 8
- 8
- Word Based Chinese Semantic Role Labeling with Semantic Chunking8
The experiments have shown that the semantic chunking based method outperforms previously best-reported results on Chinese SRL, if the word segmentation and part-of-speech (POS) tagging are both correct.
- FaithLens: Detecting and Explaining Faithfulness Hallucination7
FaithLens is introduced, a cost-efficient and effective faithfulness hallucination detection model that can jointly provide binary predictions and corresponding explanations to improve trustworthiness and outperforms advanced models such as GPT-5.2 and o3.2.
- A Progressive Learning Approach to Chinese SRL Using Heterogeneous Data7
A novel progressive learning model is proposed which augments the Progressive Neural Network with Gated Recurrent Adapters and can accommodate heterogeneous inputs and effectively transfer knowledge between them, to improve Chinese SRL by using heterogeneous corpora together.
- Improving the Robustness of Distantly-Supervised Named Entity Recognition via Uncertainty-Aware Teacher Learning and Student-Student Collaborative Learning6
This work proposes Uncertainty-Aware Teacher Learning that leverages the prediction uncertainty to reduce the number of incorrect pseudo labels in the self-training stage, and Student-Student Collaborative Learning that allows the transfer of reliable labels between two student networks instead of indiscriminately relying on all pseudo labels from its teacher.
- 6
- 6
- 6
- 6
- A Soft Label Strategy for Target-Level Sentiment Classification6
A history-based soft labeling model is proposed to measure the possibility of a context word as an opinion word, and a convolution layer is applied to extract local active features, and positional weights are introduced to take relative distance information into consideration.
- Learning Word Sense Embeddings from Word Sense Definitions6
Experimental results on word similarity tasks and a word sense disambiguation task show that word sense embeddings produced by the proposed approach are of high quality.
- Fast Semantic Role Labeling for Chinese Based on Semantic Chunking6
The experiments have shown that the semantic chunking based method outperforms previously best-reported results on Chinese SRL and saves a large amount of time.
- 6
- VeCAF: Vision-language Collaborative Active Finetuning with Training Objective Awareness5
VeCAF optimizes a parametric data selection model by incorporating the training objective of the model being tuned, and guides the PVM towards the performance goal with improved data and computational efficiency.
- 5
- A Span-based Target-aware Relation Model for Frame-semantic Parsing5
This work reformulates the FSRL task into a target-aware relation classification task and proposes a novel and lightweight jointly learning framework that simultaneously processes three subtasks of FSP, including frame identification, argument identification, and role classification.
- Interactive Neural Network: Leveraging Part-of-Speech Window for Aspect Term Extraction (Student Abstract)5
A novel model is proposed which combines POS and word features in an interactive way, and makes full use of the POS tags of adjacent words by POS window, and proves the effectiveness of the model.
- Recognizing Textual Entailment via Multi-task Knowledge Assisted LSTM5
A deep neural network architecture called Multi-task Knowledge Assisted LSTM (MKAL) is proposed, which aims to conduct implicit inference with the assistant of KB and use predicate-to-predicate attention to detect the entailment between predicates.
- Semantic Role Labeling Using Recursive Neural Network5
A recursive neural network model is designed for semantic role labeling which is not dependent on lots of rich hand-designed features and able to add many shallow features.
- SSMT:A Machine Translation Evaluation View To Paragraph-to-Sentence Semantic Similarity5
The system SSMT measuring the semantic similarity between a paragraph and a sentence submitted to the SemEval 2014 task3: Cross-level Semantic Similarity is presented, which is straightforward in intuition and easy in implementation.
- Chinese Multi-word Chunks Extraction for Computer Aided Translation5
hybrid method proved to be a suitable method for selecting multi-word chunks, it has considerably improved the precision of the extraction which is much higher than that of purely statistical method.
- SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows?4
SaaS-Bench is introduced, a benchmark built on 23 deployable SaaS systems across six professional domains, containing 106 tasks grounded in realistic work scenarios, showing that representative LLM-based agents struggle on SaaS-Bench, exposing limitations in planning, state tracking, cross-application context maintenance, and error recovery.
- 4
- 4
- 4
- Discovering Concept-Level Event Associations from a Text Stream4
This work investigates the importance and challenge of this task and proposes a novel solution by using event sequential patterns, which can discover important event associations implicitly expressed.
- 3
- 3
- 3
- SeRI: A Dataset for Sub-event Relation Inference from an Encyclopedia3
A novel dataset called SeRI (Sub-event Relation Inference), which includes 3,917 event articles from English Wikipedia and the annotations of their sub-events and can be used for training or evaluating a model that mines sub-event relation from encyclopedia-style texts.
- Improving Chinese Semantic Role Labeling with English Proposition Bank3
This paper introduces a two-pass approach to do Chinese SRL with a Recurrent Neural Network (RNN) model and uses English Proposition Bank (EPB) to improve the performance ofChinese SRL.
- Distantly Supervised Neural Network Model for Relation Extraction3
A novel approach to alleviate the problem of distant supervision with representation learning in the framework of deep neural network that constructs the more powerful mention level representation by tensor-based transformation and further learns the entity pair level representation which aggregates and denoises the features of associated mentions.
- A Refined HDP-Based Model for Unsupervised Chinese Word Segmentation3
A refined Hierarchical Dirichlet Process (HDP) model for unsupervised Chinese word segmentation is proposed and it is shown that the initial segmentation state for HDP model plays a very important role in model performance.
- Chinese Semantic Role Labeling Based on Semantic Chunking3
A semantic chunking based method is established which changes the SRL task from the traditional "parsing-semantic role identification-Semantic role classification" process into a simple "semantic chunk identification- Semantic chunk classification" pipeline.
- 3
- 3
- 3
- RoadmapBench: Evaluating Long-Horizon Agentic Software Development Across Version Upgrades2
A systematic evaluation on thirteen frontier models finds that even the strongest, Claude-Opus-4.7, resolves only 39.1% of tasks, while the weakest achieves merely 5.2%, suggesting that long-horizon software development remains a largely unsolved problem.
- LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning2
This paper proposes LongAct, a strategy that shifts from uniform to saliency-guided sparse updates, which achieves an approximate 8% improvement on LongBench v2 and enhances generalization on the RULER benchmark.
- 2
- 2
- 2
- A Two-Stage Method for Chinese AMR Parsing2
A two-stage method to conduct Chinese AMR Parsing with alignment generation, which includes Concept-prediction and Relation-Prediction stages is proposed, which achieves 0.7756 and 0.7074 Align-Smatch F1 scores on the CAMR 2.0 test set and the blind-test set of CAMRP-2022 individually.
- 2
- 2
- An Anchor-Based Automatic Evaluation Metric for Document Summarization2
A new protocol for designing reference-based metrics that require the endorsement of source document(s) is considered and an anchored ROUGE metric fixing each summary particle on source document, which bases the computation on more solid ground is proposed.
- Fine-grained Coordinated Cross-lingual Text Stream Alignment for Endless Language Knowledge Acquisition2
Experiments show this approach not only outperforms previous approaches on bilingual lexicon extraction from coordinated text streams but also can harvest high-quality alignments from large amounts of streaming data for endless language knowledge mining, which makes it promising to be a new paradigm for automatic language knowledge acquisition.
- A Research on Construction of Knowledge Fusion Network in Chinese and English Languages2
A prototype of Chinese-English lexical and encyclopedic knowledge fusion network CELK-Net is created by an automatic mapping algorithm inspired by Babelnet, a localized bilingual semantic network with large coverage in Chinese language.
- Improved Graph-Based Dependency Parsing via Hierarchical LSTM Networks2
A neural graph-based dependency parsing model which utilizes hierarchical LSTM networks on character level and word level to learn word representations, allowing the model to avoid the problem of limited-vocabulary and capture both distributional and compositional semantic information.
- 2
- Multi-phase Word Sense Embedding Learning Using a Corpus and a Lexical Ontology.2
This work proposes a multi-phase word sense embedding learning method which utilizes both a corpus and a lexical ontology to learn one embedding per word sense.
- One Tense per Scene: Predicting Tense in Chinese Conversations2
A set of novel sentence-level features are proposed using rich linguistic resources and a new hypothesis of “One tense per scene” is proposed to incorporate scene-level (global) evidence to enhance the performance.
- Semantic Role Classification Based on Peking University Chinese NetBank2
The experiments have shown that the method and the features have good performance on the new corpus and compared to the PropBank, some features play crucial roles in classification on thenew corpus.
- 2
- Maximum Entropy-Based Semantic Role Classification2
This paper focuses on the semantic role classification (SRC), one key step of SRL, and introduces some new features and explores the inter-dependence of the semantic roles.
- 2
- Rethinking Semantic Parsing for Large Language Models: Enhancing LLM Performance with Semantic Hints1
- 1
- 1
- 1
- 1
- 1
- 1
- Distinguishing Specific and Daily Topics1
This paper investigates several numeric features that describe useful statistical information for this task, and proposes a novel Bayesian model for distinguishing specific and daily topics from a collection of documents based on documents’ content.
- 1
- 1
- –
- –
- –
- –
- CCAgent: Coordinating Collaborative Data Scaling for Operating System Agents via Web3–
CCAgent Net is introduced, a system that coordinates and incentivizes internet users to contribute to scaling OS agent datasets and GUI-Pipe is proposed, an automated data post-processing pipeline that evaluates, filters, and transforms raw user-uploaded OS data into trainable image-instruction-answer data.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Soap: Soaking Capacity Optimization for Multi-Document Summarization–
This paper considers the MDS task as an optimization problem with a novel measure named soaking capacity being the objective function and proposes a greedy algorithm named Soap by adopting a surrogate of soaking capacity to accelerate the computation.
- –
- –
- –
- Multi-phase Word Sense Embedding Retrofitting with Lexical Ontology–
This work proposes a multi-phase word sense embedding retrofitting retrofitting method which utilizes a lexical ontology to learn one embedding per word sense.
- Graph-Based Dependency Parsing with Recursive Neural Network–
This paper extends recursive neural network into dependency parsing, allowing it to efficiently represent the whole sub-tree context and rich structural information for each node, and proposes a heuristic search procedure for decoding.
- A Neural Network Based Translation Constrained Reranking Model for Chinese Dependency Parsing–
A simple yet effective translation constrained reranking model that outperforms the state-of-the-art monolingual Graph/Transition-based parsers by a large margin (UAS).
- –
- Chinese word segmentation model using bootstrapping–
The proposed approach to use χ 2 statistics to increase the OOV recall and use bootstrapping strategy to increase the overall F score does help and both of the OOV re-call and the overall F score are improved.
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.