Jens Lehmann · @jenslehmann
113 followers · 16 posts · Server sigmoid.social

At in Toronto, we present our work on directly retrieving facts from a knowledge graph without explicit entity & relation linking. Work with Jinheon Baek, Alham Fikri and Sung Ju Hwang.

Paper link: arxiv.org/pdf/2305.12416.pdf

#acl2023nlp #questionanswering #LLMs #KnowledgeGraphs

Last updated 1 year ago

SEMANTiCS Conference · @semantics
59 followers · 57 posts · Server sigmoid.social

📢📢 NEW BLOG POST FROM SPONSOR

We interviewed The QA Company, and asked them a few questions regarding their main products, their role in the Semantic Web Community, and about their vision.

Read the interview here: semantics-org.github.io/page/n



#semanticsconf #questionanswering

Last updated 1 year ago

IDHN · @idhn
62 followers · 347 posts · Server fedihum.org

Updates on the 'an Shared tasks (Qur'an QA 2023) organized at WANLP 2023 Conference on December 7th, 2023 are available here:

sites.google.com/view/quran-qa

You can find information on the two tasks and register until August 10th, 2023 on the website.

#qur #questionanswering

Last updated 1 year ago

Hobson Lane · @hobs
778 followers · 2326 posts · Server mstdn.social

@msh
Not true. All the say otherwise. You have to look past the hyped to the bread and butter BERT and BART models, but the trend is undeniable:

paperswithcode.com/area/natura

You name an NLP problem and there's an LLM that is now better at it than the average human. Not so 2 yrs ago. Times they are a change'n.
@futurebird

#vqa #ner #generation #translation #questionanswering #summarization #retrieval #classification #llms #benchmarks

Last updated 1 year ago

Sören Auer 🇪🇺🇺🇦 · @soeren_auer
406 followers · 357 posts · Server mstdn.social

Everybody talks about and . A task they fail big is described in this article in Nature
@SciReports@twitter.com: nature.com/articles/s41598-023

"The SciQA Scientific Benchmark for Scholarly Knowledge" with
@OliverKarras@twitter.com,
@markusstocker@twitter.com,
@iamYaserJ@twitter.com et al.

If you think you can develop a QA system performing better than plain LLMs consider participating in Scholarly QALD
@iswc_conf@twitter.com

kgqa.github.io/scholarly-QALD-

#questionanswering #chatgpt #llms

Last updated 1 year ago

Yañn Y. Nicolas · @071625348
65 followers · 245 posts · Server mamot.fr

RT @mommi84@twitter.activitypub.actor

It is now possible to query RDF files using natural language with @gpt_index@twitter.activitypub.actor! Just plug your @OpenAI@twitter.activitypub.actor API key to enable explainable complex question answering powered by GPT-3. Supports RDF and RDFS vocabularies.

🐦🔗: twitter.com/mommi84/status/163

#rdf #knowledgegraph #questionanswering #nlproc #xai #ai #gpt3 #llm

Last updated 2 years ago

Benjamin Han · @BenjaminHan
264 followers · 384 posts · Server sigmoid.social

(2/) Key points:

* Human experts can detect ChatGPT-generated answers much more reliably than human amateurs (screenshot 1).

* Slightly more than half of the times human experts found ChatGPT-generated answers are helpful (screenshot 1).

* ChatGPT is more helpful in finance and psychology, and least in medical (screenshot 1).

* RoBERTa-based classifier can reliably detect ChatGPT-generated answers(96+ in F1; screenshot 2).

#ai #questionanswering #nlp #nlproc #deeplearning #paper

Last updated 2 years ago

Benjamin Han · @BenjaminHan
261 followers · 380 posts · Server sigmoid.social

(1/) Interesting paper comparing how distinguishable the answers from and how helpful they are compared to those from humans when the same set of questions (screenshots) are asked.

Biyang Guo, Xin Zhang, Ziyuan Wang, Minqi Jiang, Jinran Nie, Yuxuan Ding, Jianwei Yue, and Yupeng Wu. 2023. “How Close Is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection.” arXiv [cs.CL]. arXiv. arxiv.org/abs/2301.07597

#chatgpt #ai #questionanswering #nlp #nlproc #deeplearning #paper

Last updated 2 years ago

Phil Cowans · @philcowans
4 followers · 67 posts · Server universeodon.com

There's plenty of discussion of using such as for and , but I don't think I've seen a quantitative comparison with more traditional models using a standard benchmark. Is there a paper somewhere covering this?

#llms #chatgpt #informationretrieval #questionanswering

Last updated 2 years ago

Jake Mannix · @pbrane
211 followers · 151 posts · Server sigmoid.social

@astatide yeah there are good ways to use these models, but trying to generate when you want to search is definitely doing it wrong.

Generative models are drunk geniuses - they've seen all the knowledge in the world! But they're also drunk af. Like a modern oracle of Delphi, kinda!

But yes, search is: query understanding, then broad retrieval, then ranking, and finally returning _references_ which you can consult.

Even verging into requires care, to avoid spewing nonsense.

#questionanswering

Last updated 2 years ago

Jeremiah Milbauer · @jerelev
17 followers · 16 posts · Server sigmoid.social

Hello world! Here's my

I’m an student at and a Student Researcher at , interested in for {Language, Knowledge, Culture, Society}

Currently I work on with applications to , , , & .

I'm often inspired by , , , etc -- long term I'd like to study & build interactive that supports discovery, agency, access, community, and trust.

#Introduction #nlp #phd #CMU #google #ai #nlproc #computationalSocialScience #ScienceOfScience #questionanswering #ScholarlyNLP #libraries #wikis #memex #KnowledgeAugmentation

Last updated 2 years ago

Jeremiah Milbauer · @jerelev
83 followers · 35 posts · Server sigmoid.social

Hello world! Here's my

I’m an student at and a Student Researcher at , interested in for {Language, Knowledge, Culture, Society}

Currently I work on with applications to , , , & .

I'm often inspired by , , , etc -- long term I'd like to study & build interactive that supports discovery, agency, access, community, and trust.

#Introduction #nlp #phd #CMU #google #ai #nlproc #computationalSocialScience #ScienceOfScience #questionanswering #ScholarlyNLP #libraries #wikis #memex #KnowledgeAugmentation

Last updated 2 years ago