| 1 |
Grounding Multilingual Multimodal LLMs With Cultural Knowledge |
提出文化知识驱动的多语言多模态LLM以解决文化差距问题 |
large language model multimodal |
|
|
| 2 |
ALOPE: Adaptive Layer Optimization for Translation Quality Estimation using Large Language Models |
提出ALOPE框架以提升机器翻译质量估计 |
large language model |
|
|
| 3 |
"Pull or Not to Pull?'': Investigating Moral Biases in Leading Large Language Models Across Ethical Dilemmas |
评估大型语言模型在伦理困境中的道德偏见 |
large language model |
|
|
| 4 |
Assessing and Mitigating Data Memorization Risks in Fine-Tuned Large Language Models |
提出多层隐私保护框架以解决大语言模型数据记忆风险 |
large language model |
|
|
| 5 |
DySK-Attn: A Framework for Efficient, Real-Time Knowledge Updating in Large Language Models via Dynamic Sparse Knowledge Attention |
提出DySK-Attn以解决大语言模型知识更新问题 |
large language model |
|
|
| 6 |
Omni-SafetyBench: A Benchmark for Safety Evaluation of Audio-Visual Large Language Models |
提出Omni-SafetyBench以解决音视频大语言模型安全评估问题 |
large language model |
|
|
| 7 |
CCFQA: A Benchmark for Cross-Lingual and Cross-Modal Speech and Text Factuality Evaluation |
提出CCFQA基准以解决多语言多模态事实性评估问题 |
large language model multimodal |
✅ |
|
| 8 |
Gradient Surgery for Safe LLM Fine-Tuning |
提出SafeGrad以解决LLM微调中的安全性问题 |
large language model foundation model |
|
|
| 9 |
Positional Biases Shift as Inputs Approach Context Window Limits |
提出相对输入长度分析以解决长输入中的位置信息偏差问题 |
large language model |
|
|
| 10 |
Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs |
提出ReLoc框架以解决代码生成中的效率与可扩展性问题 |
large language model |
|
|
| 11 |
Think Before You Talk: Enhancing Meaningful Dialogue Generation in Full-Duplex Speech Language Models with Planning-Inspired Text Guidance |
提出TurnGuide以解决全双工语音模型对话生成中的时序问题 |
foundation model |
✅ |
|
| 12 |
Arce: Augmented Roberta with Contextualized Elucidations for Ner in Automated Rule Checking |
提出ARCE以解决AEC领域命名实体识别问题 |
large language model |
✅ |
|
| 13 |
Adapting LLMs to Time Series Forecasting via Temporal Heterogeneity Modeling and Semantic Alignment |
提出TALON框架以解决时间序列预测中的异质性问题 |
large language model |
✅ |
|
| 14 |
Schema Lineage Extraction at Scale: Multilingual Pipelines, Composite Evaluation, and Language-Model Benchmarks |
提出自动化提取多语言企业数据管道的模式血统框架 |
large language model |
|
|
| 15 |
Reflect then Learn: Active Prompting for Information Extraction Guided by Introspective Confusion |
提出主动提示框架以解决信息提取中的混淆问题 |
large language model |
|
|