We lifted 7 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| karunraju/VQA | reimplementation | 1 of 1 |
| arya46/VQA_HieCoAtt | pwc_unofficial | 0 of 5 |
| miohana/vqa | pwc_unofficial | 0 of 1 |
| Function | Status | Where it lives |
|---|---|---|
| collate_lines | Ran | karunraju/VQA/coatt/coattention_experiment_runner.py pointer only (licence: NONE) · get_code("8f94f0e75a42bcfa") |
| build_model | Not yet run | arya46/VQA_HieCoAtt/models/arch.py code served (permissive licence) · get_code("841acde4f63da42b") |
| image_feature_extractor | Not yet run | arya46/VQA_HieCoAtt/utils/helper_functions.py code served (permissive licence) · get_code("89d2b73b092fedd6") |
| image_feature_extractor | Not yet run | miohana/vqa/utils/helper_functions.py code served (permissive licence) · get_code("ab1d505a48053f78") |
| predict_answers | Not yet run | arya46/VQA_HieCoAtt/utils/helper_functions.py code served (permissive licence) · get_code("7d9a7d3d8353ba1a") |
| predict_function | Not yet run | arya46/VQA_HieCoAtt/utils/load_pickles.py code served (permissive licence) · get_code("4a692c10164ea21a") |
| process_sentence | Not yet run | arya46/VQA_HieCoAtt/utils/helper_functions.py code served (permissive licence) · get_code("fe06a6a886106510") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
A number of recent works have proposed attention models for Visual Question Answering (VQA) that generate spatial maps highlighting image regions relevant to answering the question. In this paper, we argue that in addition to modeling "where to look" or visual attention, it is equally important to model "what words to listen to" or question attention. We present a novel co-attention model for VQA that jointly reasons about image and question attention. In addition, our model reasons about the question (and consequently the image via the co-attention mechanism) in a hierarchical fashion via a novel 1-dimensional convolution neural networks (CNN). Our model improves the state-of-the-art on the VQA dataset from 60.3% to 60.5%, and from 61.6% to 63.3% on the COCO-QA dataset. By using ResNet, the performance is further improved to 62.1% for VQA and 65.4% for COCO-QA.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1606.00061")
get_code_for_paper("1606.00061")
have("1606.00061")
Connect an agent — have() is free.