SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.12374 · EMNLP · 2022

REASTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples

Dragomir Radev, Rui Zhang, Yilun Zhao, Zhenting Qi, Linyong Nan

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 11 functions out of this paper's own repositories and ran 8 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.

RepositoryRoleRan
Yale-LILY/ReasTAP canonical 8 of 11
FunctionStatusWhere it lives
convert_to_float Ran Yale-LILY/ReasTAP/utils/wikisql_utils.py
code served (permissive licence) · get_code("cad927e3d44915d5")
prepare_sythetic_qa_data Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_pretrain_data.py
code served (permissive licence) · get_code("1c0945f96ed4ee2f")
prepare_tapex_sql_data Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_pretrain_data.py
code served (permissive licence) · get_code("a2d8a8405daff36c")
process_cell Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_table_data.py
code served (permissive licence) · get_code("f6d30e2b3dac8f72")
process_one_line Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_tapex_pretrain_data.py
code served (permissive licence) · get_code("2f1deef06584aa82")
process_single_table_data Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_table_data.py
code served (permissive licence) · get_code("fdf0b1b48289cf6e")
process_tapex_data Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_tapex_pretrain_data.py
code served (permissive licence) · get_code("cbd5a5ec1c792d33")
read_json_file Ran Yale-LILY/ReasTAP/synthetic_tableqa_generation/prepare_table_data.py
code served (permissive licence) · get_code("7c1e19fabfcb24da")
func_compute_bleu Not yet run Yale-LILY/ReasTAP/utils/logicnlg_evaluate.py
code served (permissive licence) · get_code("05d2e21fa4a83b9c")
get_reference Not yet run Yale-LILY/ReasTAP/utils/logicnlg_evaluate.py
code served (permissive licence) · get_code("e90042e475d5b366")
retrieve_wikisql_query_answer_tapas Not yet run Yale-LILY/ReasTAP/utils/wikisql_utils.py
code served (permissive licence) · get_code("e58fd367888694ce")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Reasoning over tabular data requires both table structure understanding and a broad set of table reasoning skills. Current models with tablespecific architectures and pre-training methods perform well on understanding table structures, but they still struggle with tasks that require various table reasoning skills. In this work, we develop REASTAP to show that high-level table reasoning skills can be injected into models during pre-training without a complex tablespecific architecture design. We define 7 table reasoning skills, such as numerical operation, temporal comparison, and conjunction. Each reasoning skill is associated with one example generator, which synthesizes questions over semi-structured tables according to the sampled templates. We model the table pre-training task as a sequence generation task and pretrain REASTAP to generate precise answers to the synthetic examples. REASTAP is evaluated on four benchmarks covering three downstream tasks including: 1) WIKISQL-WEAK and WIKITQ for Table Question Answering; 2) TABFACT for Table Fact Verification; and 3) LOGICNLG for Faithful Table-to-Text Generation. Experimental results demonstrate that REASTAP achieves new state-of-the-art performance on all benchmarks and delivers a significant improvement on low-resource setting. Our code is publicly available at https: //github.com/Yale-LILY/ReasTAP.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2210.12374")
get_code_for_paper("2210.12374")
have("2210.12374")

Connect an agent — have() is free.