The proceedings contain 65 papers. The topics discussed include: unsupervised term extraction for highly technical domains;a hybrid approach to cross-lingual product review summarization;augmenting operations research...
ISBN:
(纸本)9781952148255
The proceedings contain 65 papers. The topics discussed include: unsupervised term extraction for highly technical domains;a hybrid approach to cross-lingual product review summarization;augmenting operations research with auto-formulation of optimization models from problem descriptions;knowledge distillation based contextual relevance matching for e-commerce product search;accelerating the discovery of semantic associations from medical literature: mining relations between diseases and symptoms;machine translation impact in e-commerce multilingual search;ask-and-verify: span candidate generation and verification for attribute value extraction;consultation checklists: standardizing the human evaluation of medical note generation;towards need-based spoken language understanding model updates: what have we learned?;and calibrating imbalanced classifiers with focal loss: an empirical study.
The proceedings contain 827 papers. The topics discussed include: generative knowledge graph construction: a review;CdConv: a benchmark for contradiction detection in Chinese conversations;transformer feed-forward lay...
The proceedings contain 827 papers. The topics discussed include: generative knowledge graph construction: a review;CdConv: a benchmark for contradiction detection in Chinese conversations;transformer feed-forward layers build predictions by promoting concepts in the vocabulary space;learning to generate question by asking question: a primal-dual approach with uncommon word generation;graph-based model generation for few-shot relation extraction;backdoor attacks in federated learning by rare embeddings and gradient ensembling;toward unifying text segmentation and long document summarization;the geometry of multilingual language model representations;improving complex knowledge base question answering via question-to-action and question-to-question alignment;co-guiding net: achieving mutual guidances between multiple intent detection and slot filling via heterogeneous semantics-label graphs;the importance of being parameters: an intra-distillation method for serious gains;and certified error control of candidate set pruning for two-stage relevance ranking.
The proceedings contain 549 papers. The topics discussed include: privacy-preserving neural representations of text;adversarial removal of demographic attributes from text data;DeClarE: debunking fake news and false c...
ISBN:
(纸本)9781948087841
The proceedings contain 549 papers. The topics discussed include: privacy-preserving neural representations of text;adversarial removal of demographic attributes from text data;DeClarE: debunking fake news and false claims using evidence-aware deep learning;it’s going to be okay: measuring access to support in online communities;detecting gang-involved escalation on social media using context;and reasoning about action sand state changes by injecting common sense knowledge.
The proceedings contain 121 papers. The topics discussed include: optimizing entity resolution in voice interfaces: an ASR-aware entity reference expansion approach;two-tiered encoder-based hallucination detection for...
ISBN:
(纸本)9798891761667
The proceedings contain 121 papers. The topics discussed include: optimizing entity resolution in voice interfaces: an ASR-aware entity reference expansion approach;two-tiered encoder-based hallucination detection for retrieval-augmented generation in the wild;the program testing ability of large language models for code;salient information prompting to steer content in prompt-based abstractive summarization;predicting entity salience in extremely short documents;salient information prompting to steer content in prompt-based abstractive summarization;predicting entity salience in extremely short documents;don’t shoot the breeze: topic continuity model using nonlinear naive bayes with attention;and retrieval augmented spelling correction for e-commerce applications.
The proceedings contain 52 papers. The topics discussed include: FreeEval: a modular framework for trustworthy and efficient evaluation of large language models;i-code studio: a configurable and composable framework f...
ISBN:
(纸本)9798891761674
The proceedings contain 52 papers. The topics discussed include: FreeEval: a modular framework for trustworthy and efficient evaluation of large language models;i-code studio: a configurable and composable framework for integrative AI;Evalverse: unified and accessible library for large language model evaluation;medico: towards hallucination detection and correction with multi-source evidence fusion;OpenOmni: a collaborative open source tool for building future-ready multimodal conversational agents;TruthReader: towards trustworthy document assistant chatbot with reliable attribution;commentator: a code-mixed multilingual text annotation framework;and Arxiv Copilot: a self-evolving and efficient LLM system for personalized academic assistance.
The proceedings contain 1268 papers. The topics discussed include: UniGen: universal domain generalization for sentiment classification via zero-shot dataset generation;Multi-News+: cost-efficient dataset cleansing vi...
ISBN:
(纸本)9798891761643
The proceedings contain 1268 papers. The topics discussed include: UniGen: universal domain generalization for sentiment classification via zero-shot dataset generation;Multi-News+: cost-efficient dataset cleansing via LLM-based data annotation;FIZZ: factual inconsistency detection by zoom-in summary and zoom-out document;table question answering for low-resourced Indic languages;LLM-based agent society investigation: collaboration and confrontation in Avalon gameplay;when LLMs meets acoustic landmarks: an efficient approach to integrate speech into large language models for depression detection;speaking in wavelet domain: a simple and efficient approach to speed up speech diffusion model;eyes don’t lie: subjective hate annotation and detection with gaze;and ThinkingFair and Slow: on the efficacy of structured prompts for debiasing language models.
The proceedings contain 77 papers. The topics discussed include: BeautifulPrompt: towards automatic prompt engineering for text-to-image synthesis;enhancing language model with unit test techniques for efficient regul...
The proceedings contain 77 papers. The topics discussed include: BeautifulPrompt: towards automatic prompt engineering for text-to-image synthesis;enhancing language model with unit test techniques for efficient regular expression generation;a comparative analysis of task-agnostic distillation methods for compressing transformer language models;towards effective automatic debt collection with persona awareness;efficient transformer knowledge distillation: a performance review;CDD: a large scale dataset for legal intelligence research;personalized dense retrieval on global index for voice-enabled conversational systems;and Text2Topic: multi-label text classification system for efficient topic detection in user generated content with zero-shot capabilities.
The proceedings contain 52 papers. The topics discussed include: fabricator: an open source toolkit for generating labeled training data with teacher LLMs;end-to-end evaluation for low-latency simultaneous speech tran...
The proceedings contain 52 papers. The topics discussed include: fabricator: an open source toolkit for generating labeled training data with teacher LLMs;end-to-end evaluation for low-latency simultaneous speech translation;ChatReport: democratizing sustainability disclosure analysis through LLM-based tools;RaLLe: a framework for developing and evaluating retrieval-augmented large language models;VIST5: an adaptive, retrieval-augmented language model for visualization-oriented dialog;koala: an index for quantifying overlaps with pre-training corpora;Sudowoodo: a Chinese lyric imitation system with source lyrics;ConvLab-3: a flexible dialogue system toolkit based on a unified data format;FLEEK: factual error detection and correction with evidence retrieved from external knowledge;and YATO: yet another deep learning based text analysis open toolkit.
The proceedings contain 1047 papers. The topics discussed include: IAG: induction-augmented generation framework for answering reasoning questions;absolute position embedding learns sinusoid-like waves for attention b...
ISBN:
(纸本)9798891760608
The proceedings contain 1047 papers. The topics discussed include: IAG: induction-augmented generation framework for answering reasoning questions;absolute position embedding learns sinusoid-like waves for attention based on relative position;decoding the silent majority: inducing belief augmented social graph with large language model for response forecasting;fine-grained conversational decoding via isotropic and proximal search;holistic inter-annotator agreement and corpus coherence estimation in a large-scale multilingual annotation campaign;evaluating the rationale understanding of critical reasoning in logical reading comprehension;evaluating and modeling attribution for cross-lingual question answering;theory of mind for multi-agent collaboration via large language models;establishing trustworthiness: rethinking tasks and model evaluation;let’s think frame by frame with VIP: a video infilling and prediction dataset for evaluating video chain-of-thought;and understanding compositional data augmentation in typologically diverse morphological inflection.
The proceedings contain 42 papers. The topics discussed include: CogKTR: a knowledge-enhanced text representation toolkit for naturallanguage understanding;LM-Debugger: an interactive tool for inspection and interven...
ISBN:
(纸本)9781959429418
The proceedings contain 42 papers. The topics discussed include: CogKTR: a knowledge-enhanced text representation toolkit for naturallanguage understanding;LM-Debugger: an interactive tool for inspection and intervention in transformer-based language models;EasyNLP: a comprehensive and easy-to-use toolkit for naturallanguageprocessing;an explainable toolbox for evaluating pre-trained vision-language models;ELEVANT: a fully automatic fine-grained entity linking evaluation and analysis tool;a pipeline for generating, annotating and employing synthetic data for real world question answering;DeepKE: a deep learning based knowledge extraction toolkit for knowledge base population;evaluate & evaluation on the hub: better best practices for data and model measurements;and KeywordScape: visual document exploration using contextualized keyword embeddings.
暂无评论