Narrative comprehension on long stories and novels has been a challenging domain attributed to their intricate plotlines and entangled, often evolving relations among characters and entities. Given the LLM's diminished reasoning over extended context and its high computational cost, retrieval-based approaches remain a pivotal role in practice. However, traditional RAG methods could fall short due to their stateless, single-step retrieval process, which often overlooks the dynamic nature of capturing interconnected relations within long-range context. In this work, we propose ComoRAG, holding the principle that narrative reasoning is not a one-shot process, but a dynamic, evolving interplay between new evidence acquisition and past knowledge consolidation, analogous to human cognition on reasoning with memory-related signals in the brain. Specifically, when encountering a reasoning impasse, ComoRAG undergoes iterative reasoning cycles while interacting with a dynamic memory workspace. In each cycle, it generates probing queries to devise new exploratory paths, then integrates the retrieved evidence of new aspects into a global memory pool, thereby supporting the emergence of a coherent context for the query resolution. Across four challenging long-context narrative benchmarks (200K+ tokens), ComoRAG outperforms strong RAG baselines with consistent relative gains up to 11% compared to the strongest baseline. Further analysis reveals that ComoRAG is particularly advantageous for complex queries requiring global comprehension, offering a principled, cognitively motivated paradigm for retrieval-based stateful reasoning.

Identifying speakers of quotations in narratives is an important task in literary analysis, with challenging scenarios including the out-of-domain inference for unseen speakers, and non-explicit cases where there are no speaker mentions in surrounding context. In this work, we propose a simple and effective approach SIG, a generation-based method that verbalizes the task and quotation input based on designed prompt templates, which also enables easy integration of other auxiliary tasks that further bolster the speaker identification performance. The prediction can either come from direct generation by the model, or be determined by the highest generation probability of each speaker candidate. Based on our approach design, SIG supports out-of-domain evaluation, and achieves open-world classification paradigm that is able to accept any forms of candidate input. We perform both cross-domain evaluation and in-domain evaluation on PDNC, the largest dataset of this task, where empirical results suggest that SIG outperforms previous baselines of complicated designs, as well as the zero-shot ChatGPT, especially excelling at those hard non-explicit scenarios by up to 17% improvement. Additional experiments on another dataset WP further corroborate the efficacy of SIG.

This work introduces an original and practical paradigm for narrative comprehension, stemming from the characteristics that individual passages within narratives tend to be more cohesively related than isolated.Complementary to the common end-to-end paradigm, we propose a fine-grained modeling of narrative context, by formulating a graph dubbed NARCO, which explicitly depicts task-agnostic coherence dependencies that are ready to be consumed by various downstream tasks.In particular, edges in NARCO encompass free-form retrospective questions between context snippets, inspired by human cognitive perception that constantly reinstates relevant events from prior context.Importantly, our graph formalism is practically instantiated by LLMs without human annotations, through our designed two-stage prompting scheme.To examine the graph properties and its utility, we conduct three studies in narratives, each from a unique angle: edge relation efficacy, local context enrichment, and broader application in QA.All tasks could benefit from the explicit coherence captured by NARCO.

When reading a story, humans can quickly understand new fictional characters with a few observations, mainly by drawing analogies to fictional and real people they already know. This reflects the few-shot and meta-learning essence of humans' inference of characters' mental states, *i.e.*, theory-of-mind (ToM), which is largely ignored in existing research. We fill this gap with a novel NLP dataset in a realistic narrative understanding scenario, ToM-in-AMC. Our dataset consists of $\sim$1,000 parsed movie scripts, each corresponding to a few-shot character understanding task that requires models to mimic humans' ability of fast digesting characters with a few starting scenes in a new movie. We further propose a novel ToM prompting approach designed to explicitly assess the influence of multiple ToM dimensions. It surpasses existing baseline models, underscoring the significance of modeling multiple ToM dimensions for our task. Our extensive human study verifies that humans are capable of solving our problem by inferring characters' mental states based on their previously seen movies. In comparison, all the AI systems lag $>20\%$ behind humans, highlighting a notable limitation in existing approaches' ToM capabilities. Code and data are available at https://github.com/ShunchiZhang/ToM-in-AMC

Existing regionalization methods tend to be either spatial attribute- or spatial interaction-based, while real-world tasks usually involve both considerations to satisfy multiple objectives simultaneously. In this research, we propose Multilayer Network Community Detection and Kernel Extension (MNCD-KE), a two-step regionalization framework, as a feasible solution for such tasks. First, spatial attributes are embedded into attributes of nodes in a spatial interaction-defined multilayer network, and the kernel and marginal parts of the regions are determined by giving the membership value of the regionalization units to network communities. Second, the final result is obtained through a kernel extension process considering geographical constraints, including spatial contiguity, size balance, morphological regularity, and existing boundary consistency of the regions. Empirical experiments show that the proposed method yields outcomes that, in maintaining comparable performances with most baseline algorithms with either ‘attribute’ or ‘interaction’ objectives as measured by the respective criteria, simultaneously meet the dual objectives with results intuitively comprehensible. Its low computing costs and parameter adjustment flexibility make the proposed framework a convenient approach for real-world multi-objective regionalization tasks. We conclude the research with discussions on the boundary conditions for the framework to work and their relevance to city science theories, along with practical implications.

We present a new task setting for attribute mining on e-commerce products, serving as a practical solution to extract open-world attributes without extensive human intervention.Our supervision comes from a high-quality seed attribute set bootstrapped from existing resources, and we aim to expand the attribute vocabulary of existing seed types, and also to discover any new attribute types automatically.A new dataset is created to support our setting, and our approach Amacer is proposed specifically to tackle the limited supervision.Especially, given that no direct supervision is available for those unseen new attributes, our novel formulation exploits selfsupervised heuristic and unsupervised latent attributes, which attains implicit semantic signals as additional supervision by leveraging product context.Experiments suggest that our approach surpasses various baselines by 12 F1, expanding attributes of existing types significantly by up to 12 times, and discovering values from 39% new types.Our data and code can be found at https://github.com/lxucs/woam.

The estimation accuracy of first-arrival-path (FAP) delay plays a vital role in positioning performance. We investigate the limitations of traditional cross-correlation (CC) algorithms in delay estimation. Our work proposes a FAP delay estimation mechanism using conditional tabular generative adversarial network (CTGAN) assisted convolutional neural network (CNN). The mechanism uses the CC algorithm to extract the delay feature in the wireless signal as input and finally outputs the FAP delay. For communication scenarios where it is difficult to obtain a large amount of training data, we use CTGAN to assist CNN training to improve the accuracy of FAP delay estimation. A series of simulation experiments were presented to evaluate the performance of CTGAN-assisted CNN and compare it with traditional high-resolution delay estimation algorithms. The results show that CNN performs well in weak LOS signals and dense multipath situations. It can still maintain high precision in the case of insufficient data.

We target the task of cross-lingual Machine Reading Comprehension (MRC) in the direct zero-shot setting, by incorporating syntactic features from Universal Dependencies (UD), and the key features we use are the syntactic relations within each sentence. While previous work has demonstrated effective syntax-guided MRC models, we propose to adopt the inter-sentence syntactic relations, in addition to the rudimentary intra-sentence relations, to further utilize the syntactic dependencies in the multi-sentence input of the MRC task. In our approach, we build the Inter-Sentence Dependency Graph (ISDG) connecting dependency trees to form global syntactic relations across sentences. We then propose the ISDG encoder that encodes the global dependency graph, addressing the inter-sentence relations via both one-hop and multi-hop dependency paths explicitly. Experiments on three multilingual MRC datasets (XQuAD, MLQA, TyDiQA-GoldP) show that our encoder that is only trained on English is able to improve the zero-shot performance on all 14 test sets covering 8 languages, with up to 3.8 F1 / 5.2 EM improvement on-average, and 5.2 F1 / 11.2 EM on certain languages. Further analysis shows the improvement can be attributed to the attention on the cross-linguistically consistent syntactic path. Our code is available at https://github.com/lxucs/multilingual-mrc-isdg.

Numbers are essential components of text, like any other word tokens, from which natural language processing (NLP) models are built and deployed. Though numbers are typically not accounted for distinctly in most NLP tasks, there is still an underlying amount of numeracy already exhibited by NLP models. For instance, in named entity recognition (NER), numbers are not treated as an entity with distinct tags. In this work, we attempt to tap the potential of state-of-the-art language models and transfer their ability to boost performance in related downstream tasks dealing with numbers. Our proposed classification of numbers into entities helps NLP models perform well on several tasks, including a handcrafted Fill-In-The-Blank (FITB) task and on question answering, using joint embeddings, outperforming the BERT and RoBERTa baseline classification.

Second-pass rescoring is an important component in automatic speech recognition (ASR) systems that is used to improve the outputs from a first-pass decoder by implementing a lattice rescoring or n-best re-ranking. While pretraining with a masked language model (MLM) objective has received great success in various natural language understanding (NLU) tasks, it has not gained traction as a rescoring model for ASR. Specifically, training a bidirectional model like BERT on a discriminative objective such as minimum WER (MWER) has not been explored. Here we show how to train a BERT-based rescoring model with MWER loss, to incorporate the improvements of a discriminative loss into fine-tuning of deep bidirectional pretrained models for ASR. Specifically, we propose a fusion strategy that incorporates the MLM into the discriminative training process to effectively distill knowledge from a pretrained model. We further propose an alternative discriminative loss. This approach, which we call RescoreBERT, reduces WER by 6.6%/3.4% relative on the LibriSpeech clean/other test sets over a BERT baseline without discriminative objective. We also evaluate our method on an internal dataset from a conversational agent and find that it reduces both latency and WER (by 3 to 8% relative) over an LSTM rescoring model.

We target on the document-level relation extraction in an end-to-end setting, where the model needs to jointly perform mention extraction, coreference resolution (COREF) and relation extraction (RE) at once, and gets evaluated in an entity-centric way.Especially, we address the two-way interaction between COREF and RE that has not been the focus by previous work, and propose to introduce explicit interaction namely Graph Compatibility (GC) that is specifically designed to leverage task characteristics, bridging decisions of two tasks for direct task interference.Our experiments are conducted on DocRED and DWIE; in addition to GC, we implement and compare different multi-task settings commonly adopted in previous work, including pipeline, shared encoders, graph propagation, to examine the effectiveness of different interactions.The result shows that GC achieves the best performance by up to 2.3/5.1 F1 improvement over the baseline.

Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages.In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with uncertainty estimation in the process to select high-quality silver labels.Three different uncertainties are adapted and analyzed specifically for the cross lingual transfer: Language Heteroscedastic/Homoscedastic Uncertainty (LEU/LOU), Evidential Uncertainty (EVI).We evaluate our framework with uncertainties on two cross-lingual tasks including Named Entity Recognition (NER) and Natural Language Inference (NLI) covering 40 languages in total, which outperforms the baselines significantly by 10 F1 on average for NER and 2.5 accuracy score for NLI.

This paper analyzes the impact of higher-order inference (HOI) on the task of coreference resolution.HOI has been adapted by almost all recent coreference resolution models without taking much investigation on its true effectiveness over representation learning.To make a comprehensive analysis, we implement an endto-end coreference system as well as four HOI approaches, attended antecedent, entity equalization, span clustering, and cluster merging, where the latter two are our original methods.We find that given a high-performing encoder such as SpanBERT, the impact of HOI is negative to marginal, providing a new perspective of HOI to this task.Our best model using cluster merging shows the Avg-F1 of 80.2 on the CoNLL 2012 shared task dataset in English.

With the increasing awareness of heterogeneity in cancers, better prediction of cancer prognosis is much needed for more personalized treatment. Recently, extensive efforts have been made to explore the variations in gene expression for better prognosis. However, the prognostic gene signatures predicted by most existing methods have little robustness among different datasets of the same cancer. To improve the robustness of the gene signatures, we propose a novel high-frequency sub-pathways mining approach (HiFreSP), integrating a randomization strategy with gene interaction pathways. We identified a six-gene signature (CCND1, CSF3R, E2F2, JUP, RARA and TCF7) in esophageal squamous cell carcinoma (ESCC) by HiFreSP. This signature displayed a strong ability to predict the clinical outcome of ESCC patients in two independent datasets (log-rank test, P = 0.0045 and 0.0087). To further show the predictive performance of HiFreSP, we applied it to two other cancers: pancreatic adenocarcinoma and breast cancer. The identified signatures show high predictive power in all testing datasets of the two cancers. Furthermore, compared with the two popular prognosis signature predicting methods, the least absolute shrinkage and selection operator penalized Cox proportional hazards model and the random survival forest, HiFreSP showed better predictive accuracy and generalization across all testing datasets of the above three cancers. Lastly, we applied HiFreSP to 8137 patients involving 20 cancer types in the TCGA database and found high-frequency prognosis-associated pathways in many cancers. Taken together, HiFreSP shows higher prognostic capability and greater robustness, and the identified signatures provide clinical guidance for cancer prognosis. HiFreSP is freely available via GitHub: https://github.com/chunquanlipathway/HiFreSP.