Targeted Multimodal Sentiment Classification based on Coarse-to-Fine Grained Image-Target Matching
Authors: Jianfei Yu, Jieming Wang, Rui Xia, Junjie Li
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on two benchmark TMSC datasets show that our model consistently outperforms the baselines, achieves state-of-the-art results, and presents interpretable visualizations. |
| Researcher Affiliation | Academia | Jianfei Yu , Jieming Wang , Rui Xia and Junjie Li School of Computer Science and Engineering, Nanjing University of Science and Technology, China {jfyu, wjm, rxia, jj li}@njust.edu.cn |
| Pseudocode | No | The paper does not contain any pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | Yes | The source code is released at https://github.com/NUSTM/ITM. |
| Open Datasets | Yes | We construct an Image-Target Matching dataset for Image Target Relevance and Object-Target Alignment tasks. Source. Since both tasks require the annotation of targets, we construct our dataset based on a subset of one benchmark dataset for the TMSC task (i.e., TWITTER-17), which has annotated the targets by [Lu et al., 2018]. |
| Dataset Splits | Yes | Table 1: Statistic of Our Image-Target Matching Dataset. Split #Targets #Images #I-T Related #I-T Unrelated #Annotated Boxes Train 1176 600 459 717 459 Dev 588 297 254 334 254 Test 588 280 270 318 270 Total 2352 1177 983 1369 983 |
| Hardware Specification | No | The paper mentions using RoBERTa and Faster R-CNN with ResNet-101 backbone but does not specify the hardware (e.g., GPU model, CPU) used for experiments. |
| Software Dependencies | No | The paper mentions using RoBERTa, Faster R-CNN, ResNet-101, and AdamW optimizer, but it does not specify version numbers for any of these software components or other libraries. |
| Experiment Setup | Yes | Specifically, we set the batch size to 32, the training epoch to 10, and λ1 and λ2 to 1 and 0.5. The learning rates for the TMSC task and the two auxiliary tasks are set to 1e-5 and 1e-6 respectively. |