Explore tweets tagged as #TableQA
Super proud of this: the authors of ReasonBERT (some of who @GoogleAI) used my PyTorch implementation of TAPAS in their research :) this was really my goal of contributing a model to @huggingface: to make AI models easily accessible, and to foster research in the TableQA area! 🙌
ReasonBERT: Pre-trained to Reason with Distant Supervision pdf: https://t.co/2039ucWzJS abs: https://t.co/NEdgtxCvsX a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid, contexts
3
29
246
We have 8 categories of tasks, all MCQs. FigQA, TableQA, and ProtocolQA are reasoning tasks that don’t require tools. LitQA, SeqQA, dbQA, and suppQA are tool-use benchmarks for literature search, database access, etc. Cloning Scenarios are non-trivial “real-world” challenges. 2/
2
1
33
We experiment Readi on KGQA and TableQA (as an Information Retrieval task). Results show that Readi outperformance previous LLM-based methods and the vanilla LLMs.
0
0
1
Heading to #EMNLP2023 next week ✈️ If you’re interested in Code Generation🧑💻don't hesitate to check out our two papers! - ODEX, A challenging benchmark with open-domain coding queries: https://t.co/3bNJmhSZ64 - API-assisted code generation for tableQA: https://t.co/NXPKnFSKDa
2
22
94
Just instantiate a TableReader from PrimeQA, specify OmniTab and get predictions on your TableQA data! It's that simple.
0
0
0
We applied TabT5 to several tasks and domains such as TableQA (WikiSQL, Zhong et al) (FinQA, Chen et al), TableToText (ToTTo, Parikh et al), SpreadSheet Formula Prediction (Enron, Chen et al) achieving new SOTA results without the need of specialised architectures. 3/4
1
0
2
One of the early works that I did ,TableQA, is now a paper on Arxiv. TableQA is a solution for querying natural language with tabular data. The work is open source and under active development. Paper: https://t.co/gA9oBZJtGg Code:
1
3
17
Visual-TableQA: Complex Table Reasoning Benchmark - 2.5K - tables with 6K QA pairs - Multi-step reasoning over visual structures - 92% human validation agreement - Under $100 generation cost
5
10
50
June FOSS Meetup is happening this Sat, June 12. Agenda- 1. tableQA: AI Tool for querying natural language on tabular data @abhijithneil 2. IndianVotingAssistant: Making it easier to vote responsibly @AvikalpGupta 3. Open discussion and FOSS Grants RSVP https://t.co/D1U7xppeE4
1
4
13
When you're cold mailing, or reaching out to people via other platforms, introduce yourself, and if possible, mention your proudest work. When I was a fresher, I was proud of a project I authored, I would introduce myself as the author of TableQA: https://t.co/LOnUkkzLEV (6/n)
1
0
11
I have written a small blog about my product tableQA over here. Do check this out! https://t.co/qcZENXeiRn
0
0
5
Due to the concise and structured nature of tables, the knowledge contained therein may be incomplete or missing, posing a significant challenge for table question answering (TableQA) systems. However, most existing datasets either overlook the challenge of missing knowledge in
0
0
1
[1/n]🔥TableBench is coming !🚀A comprehensive and complex benchmark covering 18 fields across four main TableQA categories, rigorously testing LLM's performance in complex industrial TableQA scenarios. 📈 Dive into the details: https://t.co/Pm1CWwlO7O
#AI #LLM #TableQA
1
2
6