NLP on arXiv

Automated new publication entries for #NLP. Not affiliated with arXiv.

[Roberto Iriondo ♻️ @robiriondo]

1/6 🚀📚 Proud to share the definitive guide to enhancing LLM abilities and reliability: "Building LLMs for Production" by @Whats_AI and @_LouiePeters, published by @towards_AI.

[MBZUAI @mbzuai]

We are excited to be one of the launch collaborators in using the #LLM360 framework as it aligns with our values in making AI research and progress truly transparent.

@ylecun @hmason @AndrewYNg @karpathy @drfeifei @goodfellow_ian @pabbeel @math_rachel @fchollet

[MBZUAI @mbzuai]

MBZUAI is honored to contribute the first model to the #LLM360 open-source framework by releasing its 7B parameter English large language model (LLM) named Amber.

@robiriondo @waterluffy @ericxing @AurickQ @willieneis @HongyiWang10 @ljb121002 @FajriKoto @nikhilranjan7 @haonanlp

[LLM360 @llm360]

🚀 1/7 We are thrilled to launch LLM360 — pushing the frontier of open-source & transparent LLMs!

Starting with Amber (7B) & CrystalCoder (7B), we are releasing brand new pre-trained LLMs with all training code, data, and up to 360 model checkpoints.

🔗 https://t.co/ZcLPtYQhdQ

[MBZUAI @mbzuai]

The Institute’s mandate is to advance research, nurture the next generation of AI builders, and drive transformative contributions to the UAE’s knowledge-driven economy. See the latest from MBZUAI https://t.co/m3nBmPcYGd

#AI #UAE

[Towards AI @towards_AI]

Unlock the potential of operator fusion for accelerated AI inference with @quadric_io. Their latest article reveals a 30-200% runtime performance improvement via the 'TorchInductor' compiler backend. This and much more ↓ 📈.
https://t.co/MLoUEP6BhA

[Towards AI @towards_AI]

Inception, a @G42AI company, collaborates with @MBZUAI & @cerebras to unveil Jais—a 13-billion parameter generative pre-trained transformer model designed for Arabic and English language tasks. The model aims to serve a potential user base of over 400 million Arabic speakers, addressing a substantial gap in the availability of advanced language models for this demographic. It demonstrates improved performance metrics in Arabic language tasks compared to existing models. Learn more: https://t.co/u3RQ1Lclt7

[Towards AI @towards_AI]

1/4 Looking to equip your machine with the latest models for AI tasks? @NVIDIAAI's remarkable GPU portfolio, including the GeForce RTX 4070 Ti, RTX 4080, and RTX 4090, offer unprecedented power and efficiency. 🔥🔥🔥 Discover more:
https://t.co/HinUTtS0xq

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
A Video Is Worth 4096 Tokens: Verbalize Story Videos To Understand Them In Zero Shot. (arXiv:2305.09758v1 [https://t.co/xylX3xkcXe])
https://t.co/wll02qgYVr

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Clinical Note Owns its Hierarchy: Multi-Level Hypergraph Neural Networks for Patient-Level Representation Learning. (arXiv:2305.09756v1 [https://t.co/ADR5apFNsK])
https://t.co/91es95oGRX

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
What In-Context Learning "Learns" In-Context: Disentangling Task Recognition and Task Learning. (arXiv:2305.09731v1 [https://t.co/ADR5apFNsK])
https://t.co/5PN5xjk7vT

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Generative Table Pre-training Empowers Models for Tabular Prediction. (arXiv:2305.09696v1 [cs.LG])
https://t.co/Cl71xSaXGm

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
OOD-Speech: A Large Bengali Speech Recognition Dataset for Out-of-Distribution Benchmarking. (arXiv:2305.09688v1 [https://t.co/pt4w1FD55P])
https://t.co/ibl1FVOkqd

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Soft Prompt Decoding for Multilingual Dense Retrieval. (arXiv:2305.09025v1 [https://t.co/PAbHzB5FAn])
https://t.co/VOU6l07TvZ

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and Measurements of Performance. (arXiv:2305.09022v1 [https://t.co/ADR5apFNsK])
https://t.co/ZgJ0jjwT7D

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Helping the Helper: Supporting Peer Counselors via AI-Empowered Practice and Feedback. (arXiv:2305.08982v1 [cs.HC])
https://t.co/sWP5Wrz0IT

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
An assessment of measuring local levels of homelessness through proxy social media signals. (arXiv:2305.08978v1 [https://t.co/sPLgk2UfAu])
https://t.co/CHFVrC8BRN

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Watermarking Text Generated by Black-Box Language Models. (arXiv:2305.08883v1 [https://t.co/ADR5apFNsK])
https://t.co/jK5bkuL51j

[Cohere @cohere]

1/ 📢The future of NLP education is here!

We are proud to announce the official launch of LLM University (LLMU) 🏫

🧵
https://t.co/StlVeZHY4V

[NLP on arXiv @arXiv_NLP]

#arXiv #NLP
Self-supervised Neural Factor Analysis for Disentangling Utterance-level Speech Representations. (arXiv:2305.08099v1 [https://t.co/RtOTKWOiNh])
https://t.co/r079bMAI1Q