OSS 117: Rio ne répond plus
2009
Search results
View Bing Yin 🇺🇦’s profile on LinkedIn, a professional community of 1 billion members. Ten years experience in machine learning and its application in information retrieval, as…
- 500+
- 3.6K
- Amazon
- Palo Alto, California, United States
38. 2022. Multilingual knowledge graph completion with self-supervised adaptive graph alignment. Z Huang, Z Li, H Jiang, T Cao, H Lu, B Yin, K Subbian, Y Sun, W Wang. arXiv preprint arXiv:2203.14987. , 2022. 38. 2022. Named entity recognition with small strongly labeled and large weakly labeled data.
Bing Yin. Director, Applied Science. Filters. Publications (36) Code/Dataset (1) Search and information retrieval (16) Conversational AI (14) Machine learning (9) Information and knowledge management (8) Security, privacy, and abuse prevention (1) Search (11) Data mining (8) Deep learning (8) Knowledge graphs (7) e-commerce (6) KDD 2023 (4)
People also ask
Who is Bing Yin?
What does Bing do as a scientist?
What does Bing do for a living?
What has Bing done for Amazon?
Unveiling the correlation between the catalytic efficiency and acidity of a metal-free catalyst in a hydrogenation reaction. A theoretical case study of the hydrogenation of ethene catalyzed by a superacid arising from a superhalogen. Physical Chemistry Chemical Physics.
- JINGFENG YANG∗, Amazon, USA
- 3 PRACTICAL GUIDE FOR DATA
- Remark 1
- 4 PRACTICAL GUIDE FOR NLP TASKS
- 4.1 Traditional NLU tasks
- Y LLMs
- LLMs
- Y LLMs
- LLMs Y
- LLMs
- 4.2 Generation tasks
- 4.3 Knowledge-intensive tasks
- 4.5 Miscellaneous tasks
- 4.6 Real world "tasks"
- Remark 7
- 5 OTHER CONSIDERATIONS
- Remark 8
- 5.3 Safety challenges
- 6 CONCLUSION AND FUTURE CHALLENGES
HONGYE JIN∗, Department of Computer Science and Engineering, Texas A&M University, USA TANG∗, Department of Computer Science, Rice University, USA RUIXIANG XIAOTIAN HAN∗, Department of Computer Science and Engineering, Texas A&M University, USA FENG∗, Department of Computer Science and Engineering, Texas A&M University, USA QIZHANG Amazon, USA HAOM...
In this section, we’ll be discussing the critical role that data plays in selecting appropriate models for downstream tasks. The impact of data on the models’ efectiveness starts during the pre-training stage and continues through to the training and inference stages.
LLMs generalize better than fine-tuned models in downstream tasks facing out-of-distribution data, such as adversarial examples and domain shifts. LLMs are preferable to fine-tuned models when working with limited annotated data, and both can be reasonable choices when abundant annotated data is available, depending on specific task requirements. I...
In this section, we discuss in detail the use cases and no use cases for LLMs in various downstream NLP tasks and the corresponding model abilities. And in Figure 2, we summarize all discussions into a decision flow. It can be a guide for a quick decision while facing a task.
Traditional NLU tasks are some fundamental tasks in NLP including text classification, named entity recognition (NER), entailment prediction, and so on. Many of them are designed to serve as intermediate steps in larger AI systems, such as NER for knowledge graph construction. 1As we mention in Section 1, LLMs are pretrained on large and diverse da...
Difficult Tasks Requiring scaling N (e.g. Reasoning, Emergent abilities ) Fine-tuned Models Required knowledge inconsistent with the real-word. Contexts contain enough knowledge Y Y LLMs Fine-tuned Models N
Multiple N Tasks LLMs Knowledge-intensive tasks Common NLU/ NLG tasks
Tasks with little relation to N language modelling (e.g. regression)
Creative and Complex text/code generation N Fine-tuned Models Just a few labeled data (Zero/Few-shot) Y Out-of-distribution (O.O.D) data
2 Fig. 2. The decision flow for choosing LLMs or fine-tuned models for user’s NLP applications. The decision flow helps users assess whether their downstream NLP applications at hand meet specific conditions and, based on that evaluation, determine whether Y LLMs or fine-tuned models are the most suitable choice for their applications. During the d...
Natural Language Generation broadly encompasses two major categories of tasks, with the goal of creating coherent, meaningful, and contextually appropriate sequences of symbols. The first type focuses on converting input texts into new symbol sequences, as exemplified by tasks like paragraph summarization and machine translation. The second type, "...
Knowledge-intensive NLP tasks refer to a category of tasks that have a strong reliance on background knowledge, domain-specific expertise, or general real-world knowledge. These tasks go beyond simple pattern recognition or syntax analysis. And they are highly dependent on memorization and proper utilization of knowledge about specific entities, ev...
This section explores miscellaneous tasks which cannot be involved in previous discussions, to better understand LLMs’ strengths and weaknesses.
In the last part of this section, we would like to discuss the usage of LLMs and fine-tuned models in real-world "tasks". We use the term "tasks" loosely, as real-world scenarios often lack well-formatted definitions like those found in academia. Many requests to models even cannot be treated as NLP tasks. Models face challenges in the real world f...
LLMs are better suited to handle real-world scenarios compared to fine-tuned models. However, evaluating the efectiveness of models in the real world is still an open problem. Handling such real-world scenarios requires coping with ambiguity, understanding context, and handling noisy input. Compared to fine-tuned models, LLMs are better equipped fo...
Despite LLMs are suitable for various downstream tasks, there are some other factors to consider, such as eficiency and trustworthiness. Our discussion of eficiency encompasses the training cost, inference latency, and parameter-eficient tuning strategies for LLMs. Meanwhile, our examination of trustworthiness includes robustness & calibration, fai...
Light, local, fine-tuned models should be considered rather than LLMs, especially for those who are sensitive to the cost or have strict latency requirements. Parameter-Eficient tuning can be a viable option for model deployment and delivery. The zero-shot approach of LLMs prohibits the learning of shortcuts from task-specific datasets, which is pr...
LLMs have demonstrated their extremely strong capabilities in many areas such as reasoning, knowledge retention, and coding. As they become more powerful and human-like, their potential to influence people’s opinions and actions in significant ways grows. As a result, some new safety challenges to our society should be considered and have caught lo...
Recent advances in large language models have been revolutionizing the field of natural language processing. Efectively using LLMs requires understanding their capabilities, and limitations for various NLP tasks. This work presents a practical guide to working with LLMs for downstream NLP tasks. We first discuss prominent models like GPT-style and ...
Dec 4, 2018 · Open access. Published: 04 December 2018. Bioinspired and bristled microparticles for ultrasensitive pressure and strain sensors. Bing Yin, Xiaomeng Liu, Hongyan Gao, Tianda Fu & Jun Yao....
Bing Yin 🇺🇦 on LinkedIn: Witnessed the growth of Amazon Machine Learning community. The first… Bing Yin 🇺🇦’s Post. Hiring PhD interns and fulltime to build GenAI for shopping. 5y....
