We lifted 10 functions out of this paper's own repositories and ran 7 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 |
|---|---|---|
| utahnlp/layer_augmentation | canonical | 7 of 10 |
| Function | Status | Where it lives |
|---|---|---|
| count_char_freq | Ran | utahnlp/layer_augmentation/get_char_idx.py code served (permissive licence) · get_code("f0513ead2d6b4ffc") |
| load_glove_vec | Ran | utahnlp/layer_augmentation/get_pretrain_vecs.py code served (permissive licence) · get_code("30ba2325f9cb9119") |
| load_lemma | Ran | utahnlp/layer_augmentation/conceptnet.py code served (permissive licence) · get_code("e71afe5fa6eca56e") |
| load_sent | Ran | utahnlp/layer_augmentation/constraint_preprocess.py code served (permissive licence) · get_code("4eb4fae2d442e6a9") |
| load_table | Ran | utahnlp/layer_augmentation/constraint_preprocess.py code served (permissive licence) · get_code("a23007de9c975680") |
| load_word_dict | Ran | utahnlp/layer_augmentation/get_char_idx.py code served (permissive licence) · get_code("403a59ace1d8b851") |
| register_char | Ran | utahnlp/layer_augmentation/get_char_idx.py code served (permissive licence) · get_code("569cc35c329c33da") |
| load_cache | Not yet run | utahnlp/layer_augmentation/constraint_preprocess.py code served (permissive licence) · get_code("47b31034a78a8672") |
| load_output | Not yet run | utahnlp/layer_augmentation/conceptnet.py code served (permissive licence) · get_code("a6c6363796e66bd6") |
| open_url | Not yet run | utahnlp/layer_augmentation/conceptnet.py code served (permissive licence) · get_code("64dc7f17cd644fe2") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Today, the dominant paradigm for training neural networks involves minimizing task loss on a large dataset. Using world knowledge to inform a model, and yet retain the ability to perform end-to-end training remains an open question. In this paper, we present a novel framework for introducing declarative knowledge to neural network architectures in order to guide training and prediction. Our framework systematically compiles logical statements into computation graphs that augment a neural network without extra learnable parameters or manual redesign. We evaluate our modeling strategy on three tasks: machine comprehension, natural language inference, and text chunking. Our experiments show that knowledge-augmented networks can strongly improve over baselines, especially in low-data regimes.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1906.06298")
get_code_for_paper("1906.06298")
have("1906.06298")
Connect an agent — have() is free.