Trainable performance upper bounds for image and video captioning

4 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

On Human Intellect and Machine Failures: Troubleshooting Integrative Machine Learning Systems (2016, methodology): “In addition to the captioning system, we conceptually applied our methodology to two other systems: (i) question answering with knowledge bases and web search (Yih et al. 2015), and (ii) an email-based reminder for a personal assistant.”
Generating captions without looking beyond objects (2016, result): “will argue that current error measures like BLEU are too insensitive to capture this. This work is complementary to previous work that disentangled the contribution of visual model and language model [10]. BLEU is a precision-oriented machine translation evaluation metric that measures the N-gram ov”